A regional real-time debris-flow warning system for the District of North Vancouver, Canada

Abstract

Engineered (structural) debris-flow mitigation for all creeks with elements at risk and subject to debris flows is often outside of the financial capability of the regulating government, and heavy task-specific taxation may be politically undesirable. Structural debris-flow mitigation may only be achieved over long (decadal scale) time periods. Where immediate structural mitigation is cost-prohibitive, an interim solution can be identified to manage residual risk. This can be achieved by implementing a debris-flow warning system that enables residents to reduce their personal risk for loss of life through timely evacuation. This paper describes Canada's first real-time debris-flow warning system which has been operated for 2 years for the District of North Vancouver. The system was developed based on discriminant function analyses of 20 hydrometric input variables consisting of antecedent rainfall and storm rainfall intensities for a total of 63 storms. Of these 27 resulted in shallow landslides and subsequent debris flows, while 36 storms were sampled that did not reportedly result in debris flows. The discriminant function analysis identified as the three most significant variables: the 4-week antecedent rainfall, the 2-day antecedent rainfall, and the 48-h rainfall intensity during the landslide-triggering storm. Discriminant functions were developed and tested for robustness against a nearby rain gauge dataset. The resulting classification functions provide a measure for the likelihood of debris-flow initiation. Several system complexities were added to render the classification functions into a usable and defensible warning system. This involved the addition of various functionality criteria such as not skipping warning levels, providing sufficient warning time before debris flows would occur, and hourly adjustment of actual rainfall vs. predicted rainfall since predicted rainfall is not error-free. After numerous iterations that involved warning threshold and cancelation refinements and further model calibrations, an optimal solution was found that best matches the actual debris-flow data record. Back-calculation of the model's 21-year record confirmed that 76% of all debris flows would have occurred during warning or severe warning levels. Adding the past 2 years of system operation, this percentage increases marginally to 77%. With respect to the District of North Vancouver boundaries, all debris flows occur during Warning and Severe Warnings emphasizing the validity of the system to the area for which it was intended. To operate the system, real-time rainfall data are obtained from a rain gauge in the District of North Vancouver. Antecedent rainfall is automatically calculated as a sliding time window for the 4-week and 2-day periods every hour. The predicted 48-h storm rainfall data are provided by the Geophysical Disaster Computational Fluid Dynamics Centre at the Earth and Ocean Science Department at the University of British Columbia and is updated every hour as rainfall is recorded during a given st...

Basic features of the predictive tools of early warning systems for water-related natural hazards: examples for shallow landslides

“…This is due to the intrinsic difficulty of finding historical data sets of rainstorms and corresponding landslides occurring in a small area, with enough data to allow reliable estimation of the probability of landslide triggering during extreme (and thus rare) rainfall events. Usually, only a few landslides occur at a site during an observation period of typically some decades, so that probabilistic landslide initiation thresholds are mostly defined at regional scale, so as to have a rich data set of observed landslides (e.g., Terlien, 1998;Guzzetti et al, 2007;Jakob et al, 2012;Ponziani et al, 2012;Segoni et al, 2015;Iadanza et al, 2016). The use of physically based models of infiltration and slope stability can help in the prediction of slope response under conditions different from those actually encountered during the observation period, thus allowing the definition of site-specific landslide initiation thresholds (e.g., Arnone et al, 2011;Ruiz-Villanueva et al, 2011;Tarolli et al, 2011;Papa et al, 2013;Peres and Cancelliere, 2014;Posner and Georgakakos, 2015;Greco and Bogaard, 2016), which can be useful for carrying out stochastic predictions.…”

An acoustic emission landslide early warning system for communities in low-income and middle-income countries

“…It is more common to use warning and forecasting approaches such as simple rainfall triggers to provide regional information (e.g. Jakob et al 2012;Martelloni et al 2012;Lagomarsino et al 2013). However, these cannot be used to give meaningful, and hence credible, site-specific warnings, with regional warnings liable to be misinterpreted and/or ignored by communities.…”

Technical Note: An operational landslide early warning system at regional scale based on space–time-variable rainfall thresholds

“…At regional scale, warning systems for rainfall-induced landslides are mainly based on empirical rainfall thresholds (Keefer et al, 1987;Aleotti, 2004;Hong et al, 2005;Tiranti and Rabuffetti, 2010;Baum and Godt, 2010;Capparelli and Tiranti, 2010;Cannon et al, 2011;Floris et al, 2012;Jakob et al, 2012;Staley et al, 2013;Lagomarsino et al, 2013;Tiranti et al, 2014;Cucchi et al, 2015). A broad literature exists on empirical rainfall thresholds for the possible initiation of landslides (Guzzetti et al, 2007, and references therein), and intensity-duration (I -D) thresholds are the most popular (Caine, 1980;Guzzetti et al, 2008;and references therein).…”