A first step towards a IoT-based local early warning system for an unsaturated slope in Norway

Abstract A framework for a IoT-based local landslide early warning system (Lo-LEWS) has been proposed. Monitoring, modelling, forecasting and represent the main phases of proposed framework. In this study, first two have applied to capture hydrological behaviour natural unsaturated slope located adjacent railway track in Eastern Norway. The is monitored stability kept under frequent observation, due its steepness presence lines at toe. commercial software GeoStudio SEEP was used create calibrate model able replicate situ volumetric water content (VWC) pore pressure (PWP) regime. simulations conducted were divided into series: one with an initial calibration VWC profile (C) another no (NC). validated using Taylor diagrams, which graphically summarize how closely pattern (or set patterns) matches observations. results show that preliminary matching VWC, as well considering climate conditions vegetation, are crucial aspects response studied slope. sensitivity analysis on hydraulic conductivity permeability anisotropy ratio contributed better define input data improve best-fit result. effectiveness best simulation, back-calculating tested 3 different time periods: 6-month, 1-year, 1.25-year. can adequately real up 1-year period, recalibration needed afterward. addition, SLOPE period coupled model. Finally, calculated safety factor (FS), temperature, precipitation, PWP dataset supervised machine learning algorithm. random forest highlighted importance FS. findings presented paper be seen step towards Internet Things (IoT)-based real-time employed Lo-LEWS.