hide
Free keywords:
-
Abstract:
Remarkably, the existing multi-hazard risk models that assess the physical vulnerability of exposed assets lack validation and do not offer the disaggregated effects per hazard scenario. We track that shortcoming while relying on the assumption that locally calibrated single-hazard vulnerability models available in the literature can be reused for multi-hazard risk assessment for building portfolios. This is done through a holistic method that probabilistically harmonises such single-hazard fragility models and allows us to assess the differential and cumulated damage that are expected from residential building stocks affected by cascading hazards. This is a modular approach that is composed of: (1) exposure models that classify the buildings into individual sets of classes for each hazard; (2) their spatial aggregation onto optimal geographical units whose variable resolution is compatible with the variability of the hazard intensities; (3) the probabilistic compatibilities between such sets of building classes and the damage states within their fragility models; (4) the use of state-dependent fragility functions for the second hazard. We test this methodology on the residential building stock of Lima (Peru), a coastal mega-city. Damage distributions and direct economic loss estimates are calculated for six independent mega-thrust-earthquake scenarios (main-shock) ranging from Mw 8.5 to 9.0. Thereby, the resultant distribution of damage states after the first earthquake is probabilistically converted to the reference frame of the tsunami and is later used to assess the cumulative damage using state-dependent fragility functions. This setting allows disaggregating the direct economic losses of each individual hazard to the entire cascading sequence.
