In 2014, a major landslide moved approximately 18 million tons of sand, clay, and sediment through more than a half mile area in Oso, Washington. The slide destroyed more than 40 homes and structures, dammed the North Fork Stillaguamish River, closed State Highway 530 for over two months, and caused 43 fatalities. Preliminary assessments of the financial impact of the Oso landslide estimate that the cost of rebuilding Highway 530 will exceed $38 million, while the total cost of emergency operations following the slide reached approximately $67 million by June 2015. These estimates do not include the costs of lost homes, property, livelihoods, and lives.
Of all the natural hazards that impact the United States, landslide activity is arguably the most pervasive. While certain regions in the United States, including the Appalachian Mountains, the Rocky Mountains, and the Pacific coastal range, frequently experience severe landslides, the U.S. Geological Survey (USGS) estimates that landslides cause in excess of $1 billion in damages and contribute to 25 to 50 deaths throughout the United States each year. These events, which occur in all U.S. states and territories, include earthflow, rock falls, mudflows, and mudslides, and are often attributable to multiple causes.
Various terms are frequently used interchangeably to describe landslides. For the purposes of this article, the term landslide will be used to describe the downslope movement of soil, rock, and organic materials under the effects of gravity, and the landform that results from such movement. The debris that makes up these earthflows may vary, but the common thread is the downslope movement that naturally occurs when the landmass gives way from the stability of the slope. The USGS attributes multiple natural causes to landslide activity throughout the United States, including rainfall, snowmelt, changes in water level, stream erosion, changes in groundwater, earthquakes, and volcanic activity. These major triggering mechanisms often contribute to the destabilization of the ground, either in isolation or, more frequently, in combination, resulting in landslide activity.
While any combination of these natural causes can and does contribute to downslope activity, recent major events have led scientists, with increasing frequency, to investigate man-made triggers as a contributing factor to landslides. Human activity, such as building roads and structures without adequate grading of slopes, alteration of drainage patterns and irrigation, and deforestation, has been found to contribute to the destabilization of land and increases the likelihood of slope failures. Lynn M. Highland & Peter Bobrowsky The Landslide Handbook––A Guide to Understanding Landslides, Reston, Virginia, U.S. Geological Survey Circular 1325, 2008. In particular, deforestation––including cutting down trees, logging, clearing land for crops, and developing unstable logging roads––has been linked to certain significant landslides in the United States in recent years.
Scientists also are investigating whether, and to what extent, climate change is impacting landslide activity in storm-affected areas. Most experts agree that climate change and global warming is leading to increased precipitation rates in some regions, particularly the Northern United States. However, some studies suggest that heavy precipitation events will occur with increasing frequency across the nation, even in the Southwestern United States. See https://climate.nasa.gov/effects. Nevertheless, there is not yet a consensus on the likely impacts of these precipitation trends. Because of the multiple factors (both natural and human) that contribute to landslide activity, scientists cannot yet predict with any accuracy the extent to which increased rainfall resulting from climate change will impact slope stability or the frequency of landslides.
When a massive landslide struck the community of Steelhead Haven in Oso, Washington, the devastation was catastrophic. Scientists estimate that nearly 270 million cubic feet of debris swept down from a 650-foot-high hillside, reaching average speeds of nearly 40 miles per hour as it covered over half a mile’s worth of land. The volume of mud, debris, and material that crashed down from the hillside was so significant that the USGS estimates it would cover approximately 600 football fields, at a depth of 10 feet. See One Year Later––The Oso Landslide in Washington, U.S. Geological Survey (Mar. 16, 2015), www.usgs.gov/news/one-year-later-–-oso-landslide- washington. In the years since the slide, scientists have revisited the slope to try to determine what caused the disaster in an area with known landslide activity. Were the causes of the landslide predictable? Should scientists have expected a landslide of such magnitude?
Four months after the Oso slide, a report published by the Geotechnical Extreme Events Reconnaissance (GEER) Association, with support from the National Science Foundation (NSF), reached preliminary conclusions about the causes of the disaster. These conclusions were based on reconnaissance observations at the Oso site and other available data, including regional and local geologic conditions and eyewitness accounts. Notably, the GEER Report found that a history of landslides in the area reflected logging activity and landmass movement that most likely contributed to the destabilization of the slope. In 2006, a landslide occurred in roughly the same area that traveled almost 330 feet and blocked the North Fork Stillaguamish River. Furthermore, a review of the historic activity in the area found multiple, less-significant episodes of slide movement in that area dating back to the 1930s. See Jeffrey R. Keaton, et al., The 22 March 2014 Oso Landslide, Snohomish County, Washington (July 22, 2014). The GEER Report also concluded that an intense three-week rainfall immediately preceding the event likely played a major role in the landslide, and that it was unlikely that the landslide was triggered by seismic activity. Furthermore, while the GEER Report emphasizes that a consideration of the impacts of timber harvesting on the Oso landslide was beyond the scope of the report, it notes that earlier studies have found a connection between timber harvesting and deep-seated landslides at other locations, and that a contributing factor to the Oso slide may have been increased instability of the ground resulting from changes to groundwater runoff that occurred because of timber harvesting in areas adjacent to the Oso landslide. However, according to the GEER Report, most of the known large landslides in the Stillaguamish River Valley predate logging activities in the area.
Finally, while the GEER Report was unable to identify any formal analyses that assessed the probability of a landslide in the Stillaguamish River Valley, it referenced multiple studies that had identified the potential for a major and disastrous landslide event that would impact human safety. However, while these reports concluded that human lives and property were at risk, no study predicted the catastrophic reach of the 2014 slide. While the three major landslides preceding the Oso event covered much shorter distances––approximately 325 to 650 feet––the 2014 landslide traveled an unprecedented 2,640 feet across the Stillaguamish River Valley.
In March 2015, one year after the Oso landslide, USGS scientists also concluded that the speed and reach of the landslide was likely attributable to several factors, including the soil’s initial porosity and water content from recent significant rainfall in the area. Consistent with the conclusions of the GEER Report, USGS scientists found that the magnitude of this slide was unprecedented in the Stillaguamish River Valley and unanticipated based on previous studies of past events in the area. In fact, while scientists agree that a significant landslide that would have likely blocked the North Fork Stillaguamish River and destroyed some structures should have been expected, it was a combination of factors that led to the unforeseen extreme size, speed, and force of the 2014 disaster. The USGS analysis further noted that the scale and speed of the landslide might have been significantly less if the underlying soil and organic materials had been denser or drier. See One Year Later––The Oso Landslide in Washington.
The GEER Report was not the only study to examine the potential for catastrophic landslides in the Stillaguamish River Valley. In 1997, scientists published a report assessing the risk of landslide liability in the Stillaguamish River Valley posed by timber harvesting in adjacent areas. The 1997 Report concluded that ground stability varied inversely with annual precipitation, and recommended that landslide, precipitation, and groundwater monitoring accompany any future timber harvests approved in that area. See Daniel J. Miller & Joan Sias, Deciphering Large Landslides: Linking Hydrological, Groundwater and Slope Stability Models through GIS, 12 Hydrological Processes, 923–41 (1998). In 1999, a report to the U.S. Army Corps of Engineers also warned of the possibility that “increasing rates of activity upslope” could “increas[e] the potential for a large catastrophic failure” at the future site of the Oso landslide. See Lynne Rodgers Miller and Daniel J. Miller, Answers in Geomorphology and Land Use Planning, M2 Environmental Services, Oct. 20, 1999.
Other potential contributing factors to the Oso landslide have begun to emerge in the wake of multiple lawsuits that have been filed against the State of Washington, state actors, and the timber company that logged an area above the site of the collapse. In these suits, plaintiffs allege that the unstable slope above the Stillaguamish River Valley was aggravated by logging activity, and by the installation of a crib wall to control sediment along the Stillaguamish River. These lawsuits are not unique––in Washington, victims of landslide damage often seek to recover their losses (both human and property) through tort litigation. In some cases, however, plaintiffs find their claims blocked by governmental or tribal immunity. This article explores the common landslide liability claims that may be brought in Washington State and the extent to which state and tribal actors can shield themselves through immunity claims.
Landslide Liability in Washington State
Two of the most predominant categories of landslide liability claims in Washington State are inverse condemnation and tort liability actions. While these are by no means an exhaustive list of those causes of action that may be brought by landslide victims against state actors or third parties, they encompass the most common claims.
Article I, section 16 of the Washington Constitution states that “[n]o private property shall be taken or damaged for public or private use without just compensation having first been made.” Wash. Const. art. I, § 16. Consistent with this constitutional provision, an inverse condemnation claim under Washington law is defined as “an action alleging a governmental ‘taking’ or ‘damaging’ that is brought to recover the value of property which has been appropriated in fact, but with no formal exercise of the power of eminent domain.” Fitzpatrick v. Okanogan Cty., 169 Wash. 2d 598, 605, 238 P.3d 1129 (2010). In order to establish that inverse condemnation has taken place, plaintiffs must show: “(1) a taking or damaging (2) of private property (3) for public use (4) without just compensation being paid (5) by a governmental entity that has not instituted formal proceedings.” Id. Notably, “both intended and unintended consequences of a governmental action may constitute a taking.” Jackass Mt. Ranch, Inc. v. S. Columbia Basin Irrigation Dist., 175 Wn. App. 374, 389, 305 P.3d 1108 (2013).
In addition to inverse condemnation proceedings, victims of landslides often bring tort claims of negligence and wrongful death against state actors. In the months following the Oso landslide, the Washington State Office of Risk Management received approximately 40 tort claims filed on behalf of the estates of victims who were killed in the slide, by property owners who survived but lost all their possessions, and by insurance companies seeking reimbursement for policy payouts. Those claims alleged that multiple entities, including county agencies, state agencies, and the timber company that allowed logging in the area above the landslide, were negligent in heeding warnings about the potential for landslides in the area. See Austin Jenkins, Dozens of Oso Slide Legal Claims Filed Against Washington, KUOW, June 11,2014, www.kuow.org/post/dozens-oso-slide-legal-claims-filed-against-washington.
The Reach of Immunity
The extent to which landslide victims can recover from any actor depends, in part, on the shield of immunity that protects many state, governmental, and tribal actors. Sovereign immunity, which traditionally applies to federal and state governments and recognized Native American tribes, protects those entities from being sued without their consent––a right guaranteed by the Eleventh Amendment of the U.S. Constitution. While this protection may allow the government to fulfill its duties without fear of lawsuits, the shield of immunity also may prevent people injured by government actors from recovering damages for their injuries or losses. However, this shield is not absolute. In some instances, immunity may be penetrated; in others, the entity under protection may waive its right to immunity. As a result, some states––including Washington––have abrogated their immunity from suit.
In 1961, Washington State eliminated its sovereign immunity doctrine. See 1961 ch. 136 § 1. In its place, the state enacted Rev. Code Wash. § 4.92.090, which provides: “[t]he state of Washington, whether acting in its governmental or proprietary capacity, shall be liable for damages arising out of its tortious conduct to the same extent as if it were a private person or corporation.” Thus, the scope of (as well as any limitations on) tort recovery that may be applied to persons or corporations also apply to the state. This represents a departure from traditional state sovereign immunity: for example, the Alabama State Constitution proclaims that “the State of Alabama shall never be made a defendant in any court of law or equity.” Ala. Const. art. I, § 14. Similarly, Connecticut law states that “[n]o state officer or employee shall be personally liable for damage or injury, not wanton, reckless or malicious, caused in the discharge of his or her duties or within the scope of his or her employment. . . . ” Conn. Gen. Stat. § 4-165.
However, Washington’s waiver of immunity is limited. In 1988, the Washington Court of Appeals held that “truly discretionary governmental acts on an executive level could not be characterized as tortious and thus could not be cognizable under the statute.” Karr v. State, 53 Wash. App. 1, 7, 765 P.2d 316 (1988). In order to determine whether an action is truly discretionary, courts examine whether the challenged act (1) involves a basic governmental policy, program, or objective; (2) is essential to the accomplishment of that policy, program, or objective; (3) requires the exercise of basic policy evaluation, judgment, and expertise; and (4) whether the governmental agency involved possesses the required control or statutory authority to do or make the challenged act. Id. at 8. Thus, the state and governmental actors are shielded when acting within their authority to achieve a governmental policy or objective.
This exception was clarified in Karr v. State, when the Washington Court of Appeals determined whether Washington State was liable for wrongful death proceedings arising from the eruption of Mount St. Helens. A year after the eruption that killed nearly 60 people, the majority of whom were outside of the restricted zones surrounding the volcano in anticipation of its eruption, relatives of the decedents sued the state, alleging that the governor was negligent in her enactment of the restricted zones around the volcano. The Washington Court of Appeals concluded that the state was immune from tort liability because the governor had made a considered, discretionary policy decision to close certain areas around Mount St. Helens, even though the closures did not extend far enough to protect decedents from the volcanic activity. Id. at 10.
Although the state can claim immunity for actions taken in furtherance of a governmental policy, program, or objective, Washington courts do not extend that blanket protection to the county or state for inverse condemnation claims. In Fitzpatrick v. Okanogan County, the Washington Supreme Court determined that neither the county nor the state was entitled to statutory immunity for claims arising from constitutional grounds, such as inverse condemnation claims. Fitzpatrick v. Okanogan Cty., 169 Wash. 2d at 606. Thus, the court concluded that the county and state were not immune from liability for property damage allegedly caused by a public flood control project maintained by the county. Id.
Finally, both federal and state law grants tribal sovereign immunity to recognized Native American tribes. This immunity shields tribes from suit, unless the tribe explicitly and unequivocally waives or abrogates its protection. Santa Clara Pueblo v. Martinez, 436 U.S. 49, 59 (1978). The immunity also extends to “governmental and commercial activities,” whether those activities are conducted “on or off a reservation.” Kiowa Tribe of Okla. v. Mfg. Techs., Inc., 523 U.S. 751, 754 (1998).
Who Is Liable for the Oso Landslide?
Following the Oso landslide, four groups of plaintiffs filed lawsuits against the state, the timber company that logged the area above the collapse, and Snohomish County. These suits, which were later consolidated, were filed by survivors and family members of the 43 people who lost their lives during the landslide. The suits alleged that logging activity by a timber company and the construction of a crib wall fence (a retaining wall frequently used where streambanks are experiencing significant erosion or mass failure) to control river sediment by a local tribe increased the danger of landslides in the area. In particular, plaintiffs alleged that the crib wall fence retained and accumulated loose soils from earlier landslides, which strengthened the 2014 landslide and impacted the leading edge of the landslide as it engulfed the Steelhead Haven neighborhood. Plaintiffs further alleged that the landslide was not a natural disaster, but instead resulted from a series of “actions and inactions” by those parties. See Hal Bernton, 10 Families of Oso Slide Victims Sue State, County, Forest Owner, The Seattle Times (Oct. 27, 2014), http://old.seattletimes.com/html/localnews/2024885040_ososlidelawsuitxml.html. Thus, plaintiffs alleged that state actors were aware of the significant risk of landslide, but failed to warn residents or mitigate the danger.
In September 2016, Snohomish County was dismissed from the suits, a decision that is currently under appeal. In October 2016, the plaintiffs settled with the State of Washington and with the timber company, prior to trial, for a collective $60 million. See Gene Johnson, Families Agree to $60 Million in Landslide Settlements, USA Today (Oct. 11, 2016), www.usatoday.com/story/news/nation/2016/10/11/oso-washington-landslide-settlement/91893332. This settlement immediately followed the September 2016 discovery that experts for the state had destroyed documents and emails related to the litigation––in October 2016, a judge ruled that the state attorney general’s office would be required to pay sanctions.
In 2017, one day before the statute of limitations expired, a second group of plaintiffs filed an almost-identical cause of action against the same three defendants. In July 2018, those plaintiffs settled with the state and the timber company for $11.5 million. See Associated Press, Second Group of Oso Landslide Families Settles for $11.5 Million, The Seattle Times, (July 22, 2018; updated July 23, 2018), www.seattletimes.com/seattle-news/second-group-of-oso-slide-families-settle-for-11-5-million.
The plaintiffs’ suits were ultimately not the only litigation to determine liability for the Oso landslide. In 2016, the Stillaguamish Tribe of Indians (the Tribe) filed a lawsuit against the State of Washington, seeking a ruling that the Tribe would not be required to indemnify the state for claims arising out of the Oso landslide. The action arose out of a 2005 project funding agreement (the Agreement), wherein the Tribe received approximately $500,000 in federal funding to build the crib wall fence to control sediment discharge into the North Fork of the Stillaguamish River in an area where sedimentation from past landslides was limiting salmon production. The Agreement also contained an indemnification clause that included a sovereign immunity waiver.
Following the lawsuits filed by the Oso landslide victims (which alleged, in part, that the Tribe’s construction of the crib wall contributed to the slide), the state sought indemnity from the Tribe and its insurer. The Tribe countered that the employee who signed the Agreement in 2005 was not authorized by Tribal law to sign the Agreement, let alone to waive sovereign immunity. The district court ultimately concluded that while the Agreement, by its own terms, expressly waived the Tribe’s sovereign immunity, that waiver was not binding on the Tribe because it was not entered into with the requisite authority. In particular, the district court noted that the Agreement was signed by an employee and non-member of the Tribe who had not been authorized to make such a waiver by the Tribe’s constitution, prior policies and practices, or any resolution. In so finding, the district court dismissed the proceeding. See Stillaguamish Tribe of Indians v. State of Washington and Robert W. Ferguson, Case No. 3:16-cv-05566-RJB (W.D. Wash. Aug. 9, 2017). This case is currently pending on appeal before the Ninth Circuit.
It is clear from this incident that Washington State’s potential liability for landslide damages is unusual. Washington is a state that witnesses a significant amount of landslide activity each year, and the abrogation of the state’s sovereign immunity rights exposes it to higher potential liability in tort actions. However, although this waiver of immunity necessarily makes the state more susceptible to suits, the courts have restricted its use, and governmental actors and other entities may not necessarily be exposed to a surfeit of liability claims. First, as discussed above, Washington State retains an exception for discretionary governmental acts on an executive level. Furthermore, the Washington Court of Appeals has clarified that in order to maintain an inverse condemnation claim, landowners must demonstrate that a governmental activity “was the direct or proximate cause of the landowner’s loss.” Halverson v. Skagit County, 139 Wn. 2d 1, 13, 983 P.2d 643 (Wash. 1999). Thus, an inverse condemnation claim against a governmental entity may survive if “the alleged damage or taking was caused by a governmental entity’s affirmative act of constructing a public project to achieve a public purpose,” or if “the alleged damage or taking of the property is reasonably necessary for the maintenance and operation of other property devoted to public use.” Jackass Mt. Ranch, Inc., 175 Wn. App. at 390. Similarly, in order to maintain a negligence claim for landslide damage, plaintiffs must demonstrate that the landowner had “notice of an alteration to the land that makes it more dangerous than if it had remained in its natural condition . . .” Price v. Seattle, 106 Wn. App. 647, 655, 24 P.3d 1098 (2001). Thus, “the duty of possessors of land to prevent landslides is limited . . . to situations where the possessor of the land has actual or constructive notice of a hazard produced by an alteration to the natural condition of the land.” Id. at 656. In such an instance, that duty includes not only a duty to warn, but a duty to “take timely corrective measures.” Id.
Notably, Washington’s abrogation of its sovereign immunity is not unique. In 1962, Alaska abolished its sovereign immunity and allowed state liability for tort actions with limited exceptions. Like Washington, Alaska grants limited immunity for actions taken by the state or a state employee, including for actions taken during the execution of a statute or in the case of a discretionary function or duty on the part of a state agency or employee. See Alaska Stat. § 09.50.250. Similarly, Georgia has waived sovereign immunity for the torts of its state officers and employees while acting within the scope of their employment. See O.C.G.A. § 50-21-23. This waiver does not extend, however, to acts or omissions by state employees during the execution of a statute, regulation, or rule, nor for the exercise (or failure to exercise) a discretionary function. See O.C.G.A. § 50-21-24. Other states, like Hawaii and Idaho, mandate that state and governmental employees are liable for tort actions arising out of the scope of their employment except, again, in limited circumstances like acts performed during the execution of a statute or a discretionary duty. See HRS §§ 662-2, 662-15; Idaho Code, §§ 6-903, 904.
While each natural disaster risk––landslides or otherwise––varies from state to state, the Oso suits may serve as an example across the nation. Victims of these disasters are turning to state actors to recover their losses and to assign responsibility for monitoring and remediation. Whether or not state actors are able to detect landslide risks, let alone prevent them, remains to be seen: In the absence of a single or significant contributing factor that was detectable and preventable, liability lawsuits will face an uphill battle. However, states and local governments can take proactive measures to defend tort and inverse condemnation claims. These measures include increased oversight and regulation of development and deforestation activities, risk assessments and ground monitoring, and preemptive warnings and risk assessment communications issued to communities at risk from landslide activity.
Will the State Act on Future Landslide Risks?
The large payouts resulting from the Oso landslide lawsuits open the question of how the state will approach future landslide risks. Studies of the causes of the Oso landslide have recommended future preventative measures that include more vigorous groundwater monitoring, a more robust warning system, and higher standards and monitoring for deforestation activity in at-risk areas. Whether these approaches will be adopted in any significant manner remains to be seen.
In January 2018, a slow-moving landslide forced evacuations and geological assessments as officials prepared for the potential collapse of Rattlesnake Ridge in Central Washington. In 2018, the Washington State Department of Natural Resources estimated that the 20-acre landmass is moving at a rate of approximately 1.5 feet per week, in a southward direction. See Rattlesnake Hills Landslide, Washington Dep’t of Natural Res., www.dnr.wa.gov/rattlesnake-hills-landslide (last visited Nov. 27, 2018). The potential slide is located on a steep slope outside of Yakima, Washington, overlooking a small community, a railroad corridor, Interstate 82, and the Yakima River.
In light of the dangers of a potential landslide, approximately 50 residents living near Rattlesnake Ridge evacuated the area in January 2018. However, as of September 2018, no major landslide event has occurred, and some residents have returned home. This has not reassured state agencies, though––the Washington State Department of Natural Resources asserts that local, state, and federal agencies; the Yakama Nation; and mine operators are working together to monitor the landslide and adjust plans as more information is gathered. Detour routes have been planned for the nearby interstate, and plans have been implemented for possible scenarios where the landslide could impact the Yakima River. In addition, state agencies have deployed monitoring systems to observe ground activity, and large containers have been placed along the southwest corner of the landslide mass to block potential rockfall. Id. Whether these warnings and cautionary measures are enough to insulate the state, state agencies, and other third parties from lawsuits in case of a major landslide event is uncertain. However, the impact of Oso is evident––proactive measures are being taken to protect the lives and property of residents, and to prevent another devastating, and possibly preventable, landslide event from occurring.