Assessment of Seismic Building Vulnerability from Space

Abstract

This paper quantitatively evaluates the suitability of multi-sensor remote sensing to assess the seismic vulnerability of buildings for the example city of Padang, Indonesia. Features are derived from remote sensing data to characterize the urban environment and are subsequently combined with in situ observations. Machine learning approaches are deployed in a sequential way to identify meaningful sets of features that are suitable to predict seismic vulnerability levels of buildings. When assessing the vulnerability level according to a scoring method, the overall mean absolute percentage error is 10.6%, if using a supervised support vector regression approach. When predicting EMS-98 classes, the results show an overall accuracy of 65.4% and a kappa statistic of 0.36, if using a naive Bayes learning scheme. This study shows potential for a rapid screening assessment of large areas that should be explored further in the future.

An Integrated Method Combining Remote Sensing Data and Local Knowledge for the Large-Scale Estimation of Seismic Loss Risks to Buildings in the Context of Rapid Socioeconomic Growth: A Case Study in Tangshan, China

“…Geiβ et al [34] IKONOS; Multi-temporal TM and ETM; nDSM Exploring the method and extent by which the seismic vulnerabilities of the buildings may be examined and assessed directly using various RS-derived building features and their environments. The in situ assessed vulnerabilities of the buildings were based on an expert scoring and EMS-98 scheme.…”

“…However, these association discussions are far from the implications of the LRCs and DPMs of buildings, which are classical expressions of seismic building vulnerabilities and were summarized based on earthquake engineering knowledge and actual earthquake damages. Some recent studies have carefully explored how remote sensing may contribute to examining and assessing the seismic structure type and/or vulnerability of a building [34,35,[39][40][41][42][43][44][45], which are the technical cores of the mainstream quantitative approaches used to determine seismic loss risks of buildings. However, uncertainties remain that require further validation.…”

“…Recently, apart from applications in post-earthquake impact assessments [24][25][26][27][28][29][30][31], many studies have addressed the application of high-resolution remote sensing technology to pre-earthquake exposure, vulnerability, and risk. In these pre-earthquake studies, the and/or -relevant issues (see Equation (1)) of various buildings were widely discussed, including various optical imagery-based specific analyses (Table 1) [32][33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50], some attempts at fusing optical imagery and LiDAR or SAR data [51][52][53] to help extract the heights of buildings, and several comprehensive reviews [54][55][56][57]. Table 1.…”

One step back for a leap forward: toward operational measurements of elements at risk

“…For instance, Taubenböck (2015) characterize the built environment with remote sensing data and retrieve specific fragility functions or damage probability matrices, respectively, for designated building types. In contrast to that, e.g., Borfecchia et al (2010), and Geiß et al (2014Geiß et al ( , 2015 combine limited in situ ground truth building inventory data with features from remote sensing and use techniques of statistical inference for a complete labelling of the residual building inventory according to relevant vulnerability levels. Similar methodological principles were exploited by Wieland et al (2012), Pittore and Wieland (2013), and Geiß et al (2016) to assess seismic vulnerability on an aggregated spatial level to allow for covering larger areas.…”

“Internet+” approach to mapping exposure and seismic vulnerability of buildings in a context of rapid socioeconomic growth: a case study in Tangshan, China

“…There are generally three major types of methods for analyzing seismic exposures, vulnerabilities, and risks: (1) the conventional and mainstream loss ratio curve (LRC) and/ or damaged probability matrix (DPM)-based methods (e.g., ATC 1985;FEMA 1999;Grünthal 1998;Inel et al 2008;Ploeger et al 2010), (2) exposure and vulnerability (E and V) or vulnerability (V) composite index-focused methods (e.g., Carreño et al 2007;Cutter et al 2003;Cutter and Finch 2008;Su et al 2007), and (3) remote sensing (RS)-based E and/or V methods (e.g., Ehrlich et al 2010Geiß et al 2014Geiß et al , 2015Geiß et al , 2016Pittore and Wieland 2013;Taubenböck et al 2009;Wieland et al 2012) (for more, please see Table 1). The conventional LRC and/or DPM-based methods can yield definite and accurate exposure, vulnerability, and risk results.…”