Assessing the Sensitivity of Seismic Loss Estimation to the Geographic Resolution of Building Exposure Model

This paper assessed the sensitivity of seismic losses to the geographic resolution of building exposure model. One of the key steps of seismic risk assessment is providing an accurate and reliable building inventory. Generally, building exposure model is derived from various sources of information with different degrees of quality and accuracy. Therefore, compilation of exposure model is a complex process that is associated with uncertainties. In this regard, selecting the most appropriate geographic resolution of building exposure model is a challenge. There is a trade-off between the accuracy of ground motion values in the centroid of grid cells and computation efficiency. On the one hand, selecting a higher resolution will result in less efficient computing. Increased grid cell size, on the other hand, will impose uncertainty on the results due to inaccuracy in estimating ground motion values in the proper location of buildings. The purpose of this study is to address this question “what is the impact of geographic resolution of exposure model on the seismic risk assessment?”. To do so, a sensitivity analysis with three distinct levels of resolution was performed in Tehran, Iran, as a case study, to evaluate the impact of exposure model resolution on estimated losses. The results showed that total damage over the region is almost insensitive to the resolution of exposure models; while, a more accurate damage map with lower standard deviation is achieved by refining resolutions. This is an important outcome that will assist researchers performing seismic risk assessment in large geographic areas, like countries or provinces, to be aware of the effects of geographic resolution of exposure model on results.