Developing a Geologically Based VS30 Site‐Condition Model for Portugal: Methodology and Assessment of the Performance of Proxies

Abstract

The inclusion of site-specific conditions is essential to adequately represent the seismic hazard and the seismic risk for a region. We acquired, gathered and organized a near surface shear-wave velocity database for Portugal, and applied a three-step methodological approach for developing a V S30 site-conditions map using extrapolation based on surface geology. The methodology includes: 1) defining a preliminary set of geologically defined units; 2) calculating the probability distribution of log V S30 for each unit; and 3) merging the units according to the results of statistical tests. The final model comprises three geologically defined units characterized by log V S30 distributions that are statistically significantly different from each other: F1-Igneous, metamorphic and old sedimentary rocks; F2-Neogene and Pleistocene formations; and F3-Holocene formations. The site conditions for F3 unit may be further refined using correlations with topographic slope based on the SRTM3 dataset. We analysed the performance site-conditions models based on correlations with exogenous data (topographic slope and surface geology analogues). The results show that the residual distributions between log V S30 values measured and estimated from those proxies are strongly biased for some geological units, emphasizing the need for acquiring regional V S data.

Re-thinking site amplification in regional seismic risk assessment

“…Furthermore, the correlation between slope and V S 30 appears highly dependent on the geological environment itself, as has been demonstrated in previous studies (e.g. Kwok et al, 2018; Thompson et al, 2014; Vilanova et al, 2018). As a result, simple application of topographically inferred V S 30 without appropriate adjustments may produce potentially erroneous amplifications, for example, predicting higher ground motions on flat terrain even in regions of older and harder surface rocks.…”

“…Although higher resolution DEMs are available at a global scale, Allen and Wald (2009) note that although these may resolve small differences in gradients for higher slopes they also introduce more variability into the slope calculations, thus reducing the confidence in the correlation with V S 30 . This observation is contrasted, however, by both Stewart et al (2017) and Vilanova et al (2018), who do report a slightly improved correlation at using higher resolution DEMs. To allow adequate comparison with the approach most widely adopted in application; however, for the V S 30 from topography used herein we maintain the same 30 arc-second DEM resolution adopted by Wald and Allen (2007).…”

“…Lemoine et al, 2012), potentially integrating other geological and/or geomorphological data (e.g. Kwok et al, 2018; Thompson et al, 2014; Thompson and Wald, 2012; Vilanova et al, 2018).…”

“…While a growing body of literature is dedicated to the improvement or calibration of V S 30 models inferred from geomorphological and/or geological proxies (e.g. Kwok et al, 2018; Thompson et al, 2014; Thompson and Wald, 2012; Vilanova et al, 2018), it is easy to lose perspective of the purpose that such models serve in terms of risk analysis, which is to characterize the site amplification itself. V S 30 serves only as a proxy for this purpose.…”

Methodology and Assessment of Proxy-Based Vs30 Estimation in Sichuan Province, China

“…Shi (2011) applied this method to establish a relational matrix between geological conditions and site categories, and completed the division of sites in China. Vilanova et al (2018) developed a near-surface shear wave velocity database for Portugal and applied a three-step methodological approach to develop a V s30 sitecondition map using extrapolation.…”