Estimation of Seismically-Induced Crest Settlement of Earth Core Rockfill Dams

Abstract

The settlement at the dam crest induced by earthquake loading is an important design parameter widely used to evaluate the seismic performance of dams. Two-dimensional nonlinear dynamic analyses were performed to calculate the dam crest settlement of earth core rockfill dams (ECRDs). The numerical model was first validated against centrifuge test measurements. Comparisons showed that the numerical model provided reasonable estimates of the measured response. A series of dynamic numerical analyses were performed for a wide range of ground motions. The calculated settlements were revealed to compare well with field recordings of dam crest settlements. The settlements were also plotted against ground motion parameters, including free-field surface peak ground acceleration (PGA), peak ground velocity (PGV), Arias intensity (Ia), and moment magnitude (M). Predictive equations for the crest settlement were also presented. It wes demonstrated that PGA provided a good estimate of the settlement, but Ia provided the most favorable fit according to the calculations. Additionally, multiple parameters models were also presented to enhance the accuracy of the predictions.

Seismic intensity measures and predictive equations for the evaluation of earthquake—induced crest settlements of earth dams

“…In addition, comparing the values of D C in Table 3 with those obtained using Equation ( 8) when H d is assumed equal to 48 m (as for the San Pietro and the Marana Capacciotti dam) or to 57 m (as for the Camastra dam) an underestimation of about 24% and an overestimation of about 30% of the crest settlement are, respectively, obtained. This observation permits tentatively concluding that for a generic earth dam, using only the height H d as input datum, Equation (8), calibrated on (the few) available field data, along with Equation ( 9), leads to a fair assessment of the earthquake induced crest settlement.…”

“…As an example, using either field data, relative to postseismic surveys, or numerical predictions, obtained through advanced finite difference or finite element analyses, relative to different types of earth dams, several studies showed the relationship between the permanent crest settlement and the peak ground acceleration or velocity (e.g., Lanzo et al, 6 Lanzo, 7 Park et al, 8 Ghaemi and Konrad, 4 Armstrong et al, 9 Regina et al 10 ), the Arias Intensity (e.g., Yule et al, 11 Sica and Pagano 5 , Brigante and Sica, 12 Pagano and Sica, 13 Lanzo et al, 6 Lanzo, 7 Park et al, 8 Armstrong et al, 9 Cascone et al, 14 Regina et al 10 ), the cumulative absolute velocity (i.e. the integral of the acceleration time-history, e.g., Lanzo et al, 6 Armstrong et al, 9 Regina et al 10 ), the bracketed (e.g., Yule et al, 11 for a threshold acceleration equal to 0.05 g) or the strong motion (Lanzo et al, 6 Armstrong et al, 9 Cascone et al, 14 Regina et al 10 ) duration, the specific energy density (i.e., the integral of the squared velocity time-history; e.g., Lanzo et al 6 ), the destructiveness potential factor (e.g., Cascone et al 14 ), the mean or the predominant period (e.g., Lanzo et al, 6 Ghaemi and Konrad 4 ) and, finally, the pseudo spectral acceleration computed for different vibration periods (e.g., Armstrong et al, 9 Regina et al 10 ).…”

“…As an alternative to these approaches some authors (e.g., Park et al, 8 Ghaemi and Konrad, 4,16 Casablanca et al, 24 Nardo et al 25 ) recently proposed empirical predictive formulas for the evaluation of the dam crest settlement calibrated on the results of numerical dynamic analyses carried out using suitable constitutive models and large sets of seismic records as input motion and relating computed displacements of the dam crest with some relevant seismic parameters.…”

“…As an example, using either field data, relative to postseismic surveys, or numerical predictions, obtained through advanced finite difference or finite element analyses, relative to different types of earth dams, several studies showed the relationship between the permanent crest settlement and the peak ground acceleration or velocity (e.g., Lanzo et al., 6 Lanzo, 7 Park et al., 8 Ghaemi and Konrad, 4 Armstrong et al., 9 Regina et al 10 11 Sica and Pagano 5 , Brigante and Sica, 12 Pagano and Sica, 13 Lanzo et al., 6 Lanzo, 7 Park et al., 8 Armstrong et al., 9 Cascone et al., 14 Regina et al 10 …”

“…As an alternative to these approaches some authors (e.g., Park et al., 8 Ghaemi and Konrad, 4,16 Casablanca et al., 24 Nardo et al 25…”