The high seismic vulnerability of the building stock in Colombia has been proven by the two main events of the last two decades: the magnitude 6.2 (Mw) Armenia earthquake of 1999 and the magnitude 5.0 (Mb) Popayán earthquake of 1983. Although both events had a relatively low magnitude, large economic losses were reported.
As in many Latin American countries, the building stock, infrastructure and population in Colombia are mainly concentrated in a few regions. Antioquia’s Department, with a medium to high seismic hazard, is the second most populated department of the country with 6,534,746 inhabitants (14% of the country’s population) and an area of 63,612 square kilometres.
A recent exposure model of the residential building stock of Antioquia was developed by EAFIT University as part of the South America Risk Assessment (SARA) project. The methodology used for the creation of the exposure model was based on available cadastral information, survey data and expert judgement. The final exposure model of Antioquia includes 125 municipalities with information about the total built-up area, number of buildings, dwellings and inhabitants, building class and replacement cost. The model indicates a total of 147 square kilometres of built-up area, 834 thousand of buildings and a total replacement cost estimated as 135×106 million of Colombian Pesos (COP). A total of 121 building classes were included in the model. These classes were defined according to the lateral load resisting system, construction materials and number of storeys. Results from the model indicate that nearly 60% of the Department building stock corresponds to unreinforced masonry structures. Unfortunately, this building typology is highly vulnerable to seismic hazard.
A more detailed exposure model was developed for the city of Medellín, capital of Antioquia, with nearly 2.5 million inhabitants. The exposure model of Medellín includes information for 350 neighbourhoods, for a total of 79 square kilometres of built-up area, 343 thousand of buildings and a total replacement cost of 76 x106 million of COP. For this particular municipality, the number of unreinforced masonry buildings increases to 70% of the building stock.
Such a large number of unreinforced masonry structures motivated the derivation of analytical fragility functions for this building typology. In-plane response of unreinforced masonry structures was defined for six building classes based on the number of stories. Forty unreinforced masonry buildings located in Medellín and surrounding municipalities were surveyed in order to gather data required for the generation of the fragility curves. The structural capacity of each building class was calculated using a simplified pushover analysis for 100 single-degree-of-freedom (SDOF) oscillators, using the Risk Modeller’s Toolkit (RMTK) developed by the GEM Foundation. Four damage states based on the SDOF displacement were considered for the fragility curve derivation: slight, moderate, extensive and collapse. Non-linear time history analyses for each SDOF oscillator were performed for 300 ground motion records. The maximum displacement of each SDOF due to each record was compared to the corresponding damage state thresholds, which allowed the derivation of a damage matrix containing the number of SDOF in each damage state. Finally, each fragility function was modelled using a cumulative lognormal distribution. As an example, a fragility function for a 3-storey unreinforced masonry building is presented below.
A plausible seismic damage scenario was assessed in order to explore the seismic risk of the unreinforced masonry building stock. The calculations were performed using the OpenQuake-engine. An earthquake with magnitude 5.5 (Mw) was simulated, considering a reverse fault rupture located in the Romeral system, as illustrated below. In addition, site conditions were considered through Vs30 values, based on the microzonation study available for the region.
Results from the assessment indicate that this event with a moderate magnitude would cause slight to moderate damage in more than 10 thousand structures, and about 5 thousand would have severe damage or collapse. This distribution of damage is presented in the plot below, as well as the mean number of collapses in the different administrative regions.
For more information about this study, please contact GEM’s risk team at firstname.lastname@example.org, or contact directly Professor Ana Beatriz Acevedo from the EAFIT University who is the main author of this study at email@example.com.