Development of the Dynamic Model of Mass Movements Threat for the Corridor of the Central Oil Pipeline S.A., “OCENSA” in the Framework of the Management of the Climate and External Forces Threat

Abstract A landslide threat analysis model is presented for the Central Oil Pipeline of Colombia, operated by OCENSA. It is part of an early warning system under development for making timely decisions to guarantee the integrity and operability of the infrastructure. The model combines susceptibility and accumulated rainfall thresholds triggering landslides with a probabilistic approach. The susceptibility analysis was performed using the weights of evidence method, considering geological, morphological, morphometric, and anthropic variables. It uses 1104 terrain instability events in a corridor of 836 × 50 kilometers. The input and output variables have a resolution of 30 × 30 meters. The validation of the model was done using the area under the receiver operating characteristic curve ROC, reaching a performance of 87.9%. The thresholds for accumulated rainfall triggering landslides are obtained using probabilistic models. Satellite monitoring of daily rainfall is conducted at 41 virtual stations using data from NASA’s Global Precipitation Measurement (GPM) mission. The stations are spaced every twenty kilometers along the pipeline axis in a 50-kilometer-wide corridor. The daily rainfall reports in the virtual stations are accumulated for different previous periods. Comparison of accumulated rainfall with thresholds allows calculating the temporal probability of landslides for different accumulation periods. The maximum temporal probabilities of instability in each sector are selected. By combining the susceptibility or spatial probability and the maximum temporal probability, the sectorized daily threat is obtained. The daily threat or space-time probability of landslides is analyzed for about 42.9 million pixels. This is represented on base cartography in five levels included in a probability range between 70% and 99.995%. The threat level at each pixel about the location of the pipeline enables prevention, preparedness, surveillance, and emergency decisions to be made to ensure the integrity and operational continuity of the system.