Exposure model dataset for Colombian residential buildings

This project provides a harmonized, block-level exposure dataset for seismic risk assessment in Colombia, focusing on residential buildings across 83 prioritized municipalities. The dataset includes geographic location, structural typology, number of stories, population, and replacement cost, offering a comprehensive representation of the built environment in areas with significant seismic hazard and population exposure. The data were generated through visual inspections, integration of existing sources, and standardization under a taxonomy specifically developed for Colombian construction practices. Quality controls ensured internal consistency, enabling reliable integration with hazard, vulnerability, and socioeconomic datasets.

Each city-level exposure model follows a standardized tabular structure including the following fields: block identifier, geographic coordinates, taxonomy (Material/System/Ductility/Height), number of stories, socioeconomic stratum, number of buildings, constructed area, physical value in million COP, occupancy, urban section (usually empty), and Vs30 soil class. Missing data are recorded as NaN. In the case of Cali, the exposure model was adapted from an external dataset that used a different taxonomy format (Use/Year_of_construction/Height/Socioeconomic_stratum). A weighted correspondence provided by the original authors was used to align it with the national taxonomy, preserving aggregate quantities and costs but without a one-to-one correspondence, which resulted in the loss of some spatial and local identifiers such as coordinates and block codes.

The dataset can be reused for probabilistic seismic risk modeling, scenario-based loss estimation, multi-hazard assessment, exposure validation studies, emergency planning, and machine-learning applications that require structured and georeferenced information about buildings. Its public availability enables reproducibility, comparability, and continuous improvement of exposure data for disaster risk reduction. A key contribution of this project is the availability of a unified, fine-resolution exposure dataset that reduces dependence on expert judgment and allows transparent evaluation of assumptions in risk studies. The methodological approach adopted here also serves as a scalable reference for future exposure development in Colombia and other regions.

The primary audience includes earthquake engineers, researchers, public institutions, urban planners, and organizations involved in disaster risk management who require reliable exposure information to support risk-informed decision-making and enhance the resilience of communities facing seismic hazards.