Re-examination of the Dynamics of the 2012 Johnsons Landing Debris Avalanche

J Aaron

Dans les comptes rendus d’articles de la conférence: GeoVancouver 2016: 69th Canadian Geotechnical Conference

Session: GEOHAZARDS - V Climate ChangeFloods & Landslides

ABSTRACT: The dynamics of the2012Johnsons Landing debris avalanche are examined using two dynamic models. The modelshave been modified to better capture the behaviour of debris moving in an undrained condition, and to explicitly accountfor the flow resistance provided by trees. It is found that simulating the debris in an undrained condition improves resultsfor this case study, and that dynamic interaction with forest has little effect on simulated impact area and velocities.R•SUM•La dynamique de l'avalanche de d†bris Johnson Landing est examin†e ‡ l'aide deuxmod…les dynamiques. Lesmod…les ont †t† modifi†s pour mieux incorporer desdebrisse d†pla—ant dans une condition non-drain†e et pourincorporer de fa—on explicite lar†sistance ‡ l'†coulement fourni par les arbres. Ce projet d†montre que simuler lesd†bris dans une condition non-drain†e am†liore les r†sultats de l'†tude de cas etque l'interaction dynamique avec lafor–t n'a peu d'effet sur la zone d'impact etles vitesses simul†es.1INTRODUCTIONExtremely rapid,flow-like landslides can travel longdistances at velocities greater than 5 m/s andsometimesas high as 30-70 m/s (Hungr et al. 2014).Quantitativerisk analysis of this class oflandslidesrequiresthat arunout analysis be performed to provideaccurateestimatesof flow depths and velocity(e.g Fell et al. 2005).Numerical runout modelsare one tool commonly used toestimate these attributes.A recent overview of variousnumericalrunout models is provided by Aaron & Hungr(2016). The numerical model thatis the subject of thiswork belongs to the class of runout models known asƒequivalent fluid models⁄.Equivalent fluid models,defined by Hungr (1995), treatthe landslide mass as a fluid whose behaviour isgoverned by internal and basal rheologies. Equivalentfluid models can account for many phenomena that havebeen observed to control the dynamics of extremely rapid,flow-like landslides,including entrainment, anisotropicinternal strength and rheology variation throughout themovement process.The parameters that controlequivalent fluid model simulations are not true materialproperties, and can only be constrained throughcalibrationback-analysis of full scale case histories. Theselection of appropriate parameters is one of the mainlimitations to the routine use of equivalent fluid models.The purpose of this paper is to examine the dynamicsof thefatalJohnsons Landing debris avalanchethatoccurred in 2012.A debris avalanche is definedbyHungret al (2014) as‹a very rapid to extremely rapid flow ofpartially or fully saturated debris on a steep slope, withoutconfinement to an established channel›.Debris is definedby Hungr et al (2014)as a low plasticity mixture of sand,gravel, cobbles and boulders.As will be discussed later, the Johnsons Landingdebris avalanche was an unusual eventthathas beenchallenging to reproduce with numerical runout models.Two previous attempts to back-analyze this event havehad to rely on ad-hoc assumptions regarding theinteraction of the debris avalanche with forest.This paperre-examinesthe dynamics of the Johnsons Landingdebris avalanche in order to rigorously test some oftheassumptions made in previous analyses. Two dynamicmodels are used to determine the appropriate rheology tosimulatethis type of event, as well as whether interactionwith forest playeda significant role in controlling thedynamics of this debris avalanche.2JOHNSONS LANDING DEBRIS AVALANCHEThe Johnsons Landing debris avalanche occurred onJuly 12th2012 about 2 km northeast of the smallcommunity of Johnsons Landing, located on KootenayLake.An image taken soon after thedebris avalancheisshowninFigure1.Thistragic event claimed four lives(Nicol et al. 2013).Asshown inFigure1,the debris avalanchewasinitially confined in a channel(Gar Creek). Just above thecommunity of Johnsons Landing,this channelhas a 70−bend. The debris avalanche avulsed from the channel atthis bend, and destroyed three homeslocatedon theJohnsons Landing bench(thelocation of the bench isshown onFigure1.)During the investigation following this tragic event,testpits were excavatedthrough the debrison the bench.Thesetestpits did not show any evidence ofotherlandslide depositson the benchsince deglaciation(Nicolet al. 2013).The 2012 eventwasthereforethe first eventto impact this bench inat least 7,700years.2.1Event DescriptionAn in-depth investigation was conductedby Nicol et al.(2013)immediately following the Johnsons Landing debrisavalanche.Previous terrain maps had identified the

Please include this code when submitting a data update: GEO2016_3884

Retrouver cet article:
Les membres de la Société canadienne de géotechnique peuvent accéder à cet article, ainsi qu'à tous les autres articles de la Conférence Géotechnique Canadienne, dans le Espace membre. Les comptes rendus d'articles sont également disponibles dans de nombreuses bibliothèques.

Citer cet article: