Seismic Response of Embankment Dams Based on Recorded Strong-Motion Data in Japan

Abstract

It is important to investigate strong-motion time series recorded at dams to understand their complex seismic responses. This paper develops a strong-motion database recorded at existing embankment dams and analyzes correlations between dam dynamic responses and ground-motion parameters. The Japan Commission on Large Dams database used here includes 190 recordings at the crests and foundations of 60 dams during 54 earthquakes from 1978 to 2012. Seismic amplifications and fundamental periods from recorded time series were computed and examined by correlation with shaking intensities and dam geometries. The peak ground acceleration (PGA) at the dam crest increases as the PGA at the foundation bedrock increases, but their ratio gradually decreases. The fundamental period broadly increases with the dam height and PGA at the foundation bedrock. The nonlinear dam response becomes more apparent as the PGA at the foundation bedrock becomes >0.2 g. The prediction models of these correlations are proposed for estimating the seismic response of embankment dams, which can inform the preliminary design stage.

Efficiency of ground motion intensity measures with earthquake-induced earth dam deformations

“…In terms of dynamic response, the two NDA models, due to the increased effects of soil nonlinearity, computed amplification ratios between the crest and base PGA that decreased with increased strong shaking, an effect that was also observed by Park and Kishida (2019). In terms of deformation response, the Lenihan Dam model computed increases in normalized crest settlement similar to that observed by Swaisgood (2014), whereas for the Anderson Dam model, substantially larger normalized crest settlements were computed because of liquefaction of portions of the embankment and foundation.…”

“…As part of the regression analyses, the variance (i.e., s 2 ln EDPj ln IM ) was estimated as the sum of squared residuals in the EDP divided by the residual degrees of Figure 7. Comparison of dynamic response (a) and deformation characteristics (b) of both dam models to the trends observed from actual embankment dams by Swaisgood (2014) and Park and Kishida (2019).…”

“…In terms of dynamic response, consider Figure 7a. There the ratio of crest-to-base PGA versus the base PGA computed from the Lenihan and Anderson dam models is compared to that measured from 60 instrumented embankment dams shaken in past earthquakes in Japan (Park and Kishida, 2019). Next, in terms of deformation characteristics, consider Figure 7b.…”

“…It should be noted that Swaisgood (2014) excluded all dams with liquefaction incidents. In Figure 7, only the observed trends from Swaisgood (2014) and Park and Kishida (2019) are shown, not the actual data points.…”

Modulus reductions of dam embankment materials based on downhole array time series

“…Figure 2b shows the corresponding FAS with the selected high-pass ( f c-hp ) and low-pass ( f c-lp ) corner frequencies. The FAS was smoothed using the function proposed by Konno and Ohmachi (1998), with b = 60 (Park and Kishida, 2019a). In Figure 2b, the slope of the entire FAS becomes flat below 0.046 Hz and deviates from the theoretical decay of 2.0 illustrated by the red dashed line (e.g.…”