Multiple estimates of soil structure at a vertical strong motion array: Understanding uncertainties from different shear wave velocity profiles Dimitrios Raptakis a, ⁎, Konstantia Makra b a Department of Civil Engineering, Aristotle University of Thessaloniki, Thessaloniki, Greece b Institute of Engineering Seismology and Earthquake Engineering (EPPO), Thessaloniki, Greece abstract article info Article history: Received 7 November 2014 Received in revised form 28 February 2015 Accepted 29 March 2015 Available online 3 April 2015 Keywords: Vertical strong motion array Shear wave velocity Earthquake data analysis Seismic prospecting Seismic noise array Gradient function We present a detailed study of investigation of various shear wave velocity, V S , profiles at the TST site of the Euroseistest test-site. We benefit from the availability of 62 V S models derived from earthquake records, conven- tional seismic prospecting, and seismic noise array measurements. Five groups of models provided from many different non-invasive and invasive methods (seismic interferometry, stress–strain analysis, annealing simula- tion, surface wave inversion, cross-hole and down-hole tests, and seismic noise array measurements) lead to av- eraged V S profiles. The estimate of V S models that we obtain differs depending on the technique used. In such cases, it becomes clear that, it is better to understand the differences and not simply compute an average. The per- centage of the observed disparity with respect to the average reference model albeit is small, becomes significant at certain depths and is associated with the existence of strong vertical discontinuities, thus introducing an un- certainty on the interface definition between the main formations. On the other hand, the use of V S profiles in ground simulation studies (especially 2D or 3D) usually implies the need to represent them as gradient functions. Testing representative generalized power law functions, we found that they fail to predict reliably the V S model for the total thickness of sediments. To overcome this, a third order polynomial function is proposed. Finally, we test the sensitivity of average V Sz index widely used in soil categorization and site amplification studies and find that all V S models, either measured or proposed, are equivalent in terms of V Sz vanishing any discrimination be- tween layering as well as models. © 2015 Elsevier B.V. All rights reserved. 1. Introduction Shear wave velocity (V S ) is deemed the most important parameter in earthquake engineering and engineering seismology studies. Its knowledge is useful in studies where prediction of site effects (e.g. Borcherdt, 1970; Aki, 1988; Bard, 1994, 1999; Chávez-García, 2011), seismic hazard assessment and Ground Motion Prediction Equations (e.g. Boore, 2004; Abrahamson et al., 2008; Douglas et al., 2009), microzonation studies and other specific studies in geotechnical engi- neering (i.e. liquefaction, soil–foundation–structure interaction and bearing capacity analyses) are needed. On the other hand, the explora- tion of the soil stiffness is a rather non-straightforward task due to rea- sons commonly related with the specific demands of the anticipated study, the non-unique evaluation of V S (shear wave velocity) or V S -1 (shear wave slowness) or G 0 (maximum shear modulus) values from different techniques, and the choice of the most adequate tool — if any — to obtain those values. Analyzing in brief these reasons, it could be said that for each of the above studies specific demands are necessary regarding the knowledge of V S values, sometimes inconsistent between each other; i.e. for bearing capacity (e.g. Turker, 2004; Tezcan et al., 2006; Tezcan and Ozdemir, 2012) and soil–foundation–structure interaction (e.g. Dobry and Gazetas, 1986; Veletsos and Prasad, 1989; Kim and Stewart, 2003) stud- ies precise seismic velocities very close to the free surface are inevitable, while for site effect studies, a rough distribution of V S in the sedimentary volume and that of bedrock could be adequate (e.g. Chávez-García and Faccioli, 2000; Makra et al., 2005). In addition, V S or G 0 estimates can be obtained with different explicit or indirect methods (e.g. Brown et al., 2002; Stephenson et al., 2005; Asten and Boore, 2005; Kuo et al., 2009; Raptakis, 2012, 2013), such as conventional seismic prospecting (Cross- and Down-Hole, SH refraction, Surface Wave Inversion), geo- technical in-situ and laboratory tests, and analysis of ambient noise measurements. Together with these techniques, there are several others based on the exploitation of earthquake recordings in vertical arrays, namely cross-correlation or seismic interferometry, stress–strain analy- sis (e.g. Elgamal et al., 1995; Assimaki et al., 2006), and annealing simu- lation approach (Satoh, 2006) as well. However, results of these techniques may differ, since each one of them explores the subsoil structure based on its hypotheses regarding the consideration of soil heterogeneities. The usual issues faced in V S es- timates are i) the resolution of the identified layering, which is strongly Engineering Geology 192 (2015) 1–18 ⁎ Corresponding author. E-mail address: raptakis@auth.gr (D. Raptakis). http://dx.doi.org/10.1016/j.enggeo.2015.03.016 0013-7952/© 2015 Elsevier B.V. All rights reserved. Contents lists available at ScienceDirect Engineering Geology journal homepage: www.elsevier.com/locate/enggeo