Earthquake Response Analysis and Evaluation for Earth-Rock Dams

principal stress of the dam body is 2534kPa, the maximum second principal stress is 2445kPa and the maximum third principal stress is 1842kPa. The maximum principal stress of the upstream and downstream dam shell occurs at the bottom of dam near the dam axis. The shear stress level of most rock-fill units are less than 0.85, no the shear failure zone appears in the dam body. It indicates that the dam is stable. This book sheds lights on recent advances in Geotechnical Earthquake Engineering with special emphasis on soil liquefaction, soil-structure interaction, seismic safety of dams and underground monuments, mitigation strategies against landslide and fire whirlwind resulting from earthquakes and vibration of a layered rotating plant and Bryan's effect. The book contains sixteen chapters covering several interesting research topics written by researchers and experts from several countries. The research reported in this book is useful to graduate students and researchers working in the fields of structural and earthquake engineering. The book will also be of considerable help to civil engineers working on construction and repair of engineering structures, such as buildings, roads, dams and monuments.

west of Port-au-Prince which is the capital city of Haiti. Approximately 230,000 people died immediately or during ensuing weeks, mostly due to acute trauma (Kenneth, 2011). The most devastating earthquake (M s =9.0) in Japan after the 1923 Great Kanto earthquake hit the eastern Japan in the afternoon of March 11, 2011. Nearly 30,000 people were killed or missed in the Tohoku earthquake and the ensuing monster tsunami (Takewaki et al., 2011). In this paper, the non-linear FEM is adopted to calculate the Wenchuan earthquake response. Then, the earthquake-resistance safety of the Bikou dam is evaluated.

Bikou earth core dam
The Bikou hydropower station is located on the Bailong River in Gansu province of China, where is about 240km away from the epicenter of Wenchuan earthquake, China, May 12, 2008 (Chen et al., 2008;Yu et al., 2009;Lin et al., 2009). The Bikou earth core rockfill dam is a 100m graduate earth-rock dam (Gu, 1989). According to the earthquake loss survey and preliminary report of the Bikou hydropower project suffered "5.12" Wenchuan earthquake, some damage appeared inside Bikou dam, but no dam slope slide and instability damaged phenomenon happened after the earthquake (Zhou et al., 2009;Pan, 2009).The normal water level of the reservoir is 707.0m, and its relative reservoir capacity is 5.21×10 8 m 3 . The dam is earth core rockfill dam with height of 101.8m, where the typical cross section of the dam is shown as in Fig.1. The ratio of the upstream slope of the dam are 1:1.8 and 1:2.3 respectively, and the downstream slope are 1:1.7, 1:1.85, 1:2.2 and 1:1.5 respectively. Its width of dam crest is 7.6 m at elevation 710.00m, and the crest length of dam is about 297.36 m. The dam is located in a roughly symmetric valley of V-type, the slope of river bank is about 50-60 degree, and the one of right bank is about 35-45 degree. The depth of overburden layer on the riverbed is about 25-34 m. The core wall in the riverbed area is built on the natural sandy gravel layer, the width of core wall crest is 4.0m at elevation 709.00-710.00m, and the ratio of the upstream slope of the core wall are 1:0.2 and 1:0.25 respectively, the section of the dam along the dam axis is shown as in Fig.2. Two concrete cut-off walls are adopted to connect the soil core wall and the bedrock. The first one built along the dam axis inserts the bedrock with the thickness 1.30m and crest the elevation 630.00m. And the second concrete one is located at the dam axis downstream 12m, inserting the bedrock with the thickness 0.80m and the crest 653.00m. The spillway buildings are made up of one spillway on the right bank, one desilting tunnel and two spillway tunnels in the left bank.

Calculation
Here,   u is nodal displacement incremental array,   R node load incremental array and   0 R the equivalent node load array of initial strain. www.intechopen.com

Constitutive model
In the static analysis, the soil materials and overburden layer materials are regarded as elastic-plastic object, where the Duncan-Zhang's nonlinear-elastic (E-B) model (Duncan & Chang, 1970) is carried out. The concrete and bedrock are regarded as linear elastic objects, which obey the generalized Hook's law.

Control equation of FEM
The dynamic equilibrium equation can be described as follows after the calculation domain is discretized by the finite element method The iterative method is adopted with the consideration of the variations of dynamic shear modulus and damping ratio in calculation process after the average dynamic shear strain change. In the iteration process the convergence criterion is Here, i G is the new shear modulus and 1  i G the last shear modulus. The maximum iteration times is taken for 6 to lest the iterative calculation into dead circulation.

Constitutive model
In the dynamic analysis (Mejia et al., 1981a(Mejia et al., , 1982bShen et al., 2006aShen et al., , 2010b, the equivalent nonlinear viscoelastic model is applied in the dynamic calculation and analysis, the soil materials and the overburden layers are assumed to be viscoelastic bodies. They reflect nonlinearity and hysteretic nature of the dynamic stress-strain relation by using of the equivalent dynamic shear modulus G and equivalent damping ratio  , which can be expressed as the function between the equivalent shear modulus and damping ratio and the dynamic shear strain amplitude. The key point of this model is to confirm the relationship between the maximum dynamic shear modulus max G and the average effective stress 0   . In this paper, the Hardin-Drnevich model (Hardin & Denevich, 1972a, 1972b is used, where the dynamic shear modulus and damping ratio can be calculated as follows Where 0   is average effective stress, a P atmospheric pressure, ' K modulus coefficient and n is modulus exponent. Here, max G , 0   and a p have the same dimension. The relation curves of dynamic shear strain  to dynamic shear modulus and damping ratio can be obtained by dynamic tri-axial test (Xenaki & Athanasopoulos, 2008;Zegha & Abdel-Ghaffar, 2009 Where  n is the normal stress on contact face,  the angle of internal friction on contact face, max  the maximum damping ratio and M , C are the test parameters.

Stability evaluation
The stress of element is used to calculate the factor of safety and evaluate its stability, thus the position of slide surface can be obtained. According to the Mohr-Coulomb criterion, the regions whose local safety factors are less than 1.0 are combined together to obtain the most dangerous multi-slip surface. The factor of safety on the surface is defined as the ratio of anti-sliding force to sliding force, and then the relationship between factors of safety and time can be obtained during the earthquake period. In this way, the antisliding stability is evaluated by stress when considering the unstable duration of earthquake.
During an earthquake, the dynamic strength of rockfill material is not always less than its static strength, at least we can assume it is equal to static strength. Taking compressive stress as positive and tensile stress as negative, after obtaining the static stress and dynamic stress under an earthquake by finite element method, the local factor of safety of an element can be calculated by the following formula Here,  is Poisson's ratio and 1  d , 3  d are dynamic stresses respectively.

Working behavior of Bikou earth-rock dam
According to the actual engineering conditions of the Bikou earth core rockfill dam, the 3-D non-linear FEM static and dynamic models of the dam are set up to calculate and analyze the stress and deformation characteristics of dam, including the deformation and stress of the dam shell, core wall, two cut-off walls, overburden layers, bedrock and so on. In order to consider the construction process, the process of filling the dam is divided into 15 levels to simulate, and the reservoir impoundment process is divided into 3 levels.

Initial stress field of dam
In order to calculate the earthquake responses of dam, the initial stress field of dam must be determined firstly. Thus, the 3-D finite element model is created to simulate the construction process of dam and reservoir impounding, and the initial stress field of dam before earthquake can be obtained.

Finite element model
According to the actual situation of the Bikou dam, the 3-D finite element model of the dam and its foundation is set up, which has simulated the geometry and material partition of the dam and its foundation. The control section super-element finite element mesh automatic division technique is adopted to generate the information of finite element model, and furthermore the super-element can be encrypted to form finite element. Based on requirements of structural characteristics, stage loading and forming the super-element grid, a set of control sections with 19 sections horizontal are selected. Then the dam and its foundation are discretized, and the supper-element grid is built up. After the super-element gird is discretized further more, the super-element gird is created whose total nodes are 31523 and total element numbers are 30087. The finite element mesh of the dam and its foundation and its core wall are shown as in Fig.3.
Select domain of the calculation model. 1) The distance from the upstream boundary to the dam axis of the river bed section is about 363.40m (approximately 1.0 times of dam height).
2) The distance from the downstream boundary to the dam toe of the river bed section is around 401.00m.
3) The distance from the left boundary to the right boundary is about 698.18m (approximately 2.0 times of dam height). 4) The vertical distance from the bottom boundary to the dam foundation surface is about 241.80 m.
The coordinate system is set up as following. The X-axis is along stream from up to down with zero at dam axis, the Y-axis is along dam axis from right bank to left bank, and the Zaxis is vertical corresponding to the elevation.
(a) dam body and foundation (b) core wall

Material parameters
The Duncan-Chang's nonlinear-elastic (E-B) model is carried out for the core wall and each district materials of the dam shell. According to indoor tests and engineering experience, the static material parameters of the dam are shown as in Table 1 and Table 2. And the parameters of the contact surface are shown as in Table 3

Displacement and stress field of dam
By use of the stage loading method to simulate the construction process of dam and reservoir impounding, the displacement field and stress field of dam under normal water level condition are obtained. Here, as examples, the displacement and stress distribution on the maximum transverse cross section of dam (Y=210m) and dam axis maximum longitudinal section of dam (X=-10m) are shown as in Fig.4 ~Fig.6.
Here, the displacement along the coordinate axis direction is positive, that is, the horizontal displacement along the flow with the direction of upstream point to the downstream is positive, the positive horizontal displacement along the dam axis direction is from the right bank to the left, and the positive vertical displacement is from bottom to up. The unit of the displacement is "mm" in some figures. The compressive stress is positive, and the tensile stress is negative. The unit of the stress is "kPa" in some figures.

Dam shell
Under normal water level condition, the maximum horizontal displacement pointing to the upstream and the downstream respectively is -289mm and 1132mm.The maximum value appears in the lower crust, close to the slope. The dam horizontal displacement distribution along the dam axis direction shows that the horizontal displacement value along dam axis direction is small. And the maximum settlement occurs in the upstream dam shell which is near the dam axis in the middle of the river. The maximum vertical displacement of the dam body is 1273mm, accounts for about 1.2% of the maximum height.
The maximum first principal stress of the dam body is 2534kPa, the maximum second principal stress is 2445kPa and the maximum third principal stress is 1842kPa. The maximum principal stress of the upstream and downstream dam shell occurs at the bottom of dam near the dam axis. The shear stress level of most rock-fill units are less than 0.85, no the shear failure zone appears in the dam body. It indicates that the dam is stable.

Core wall
The maximum horizontal displacement pointing to the downstream of the loam core wall is 812mm.The maximum horizontal displacements of core wall along dam axis direction is 181mm, pointing to the left bank. And the maximum vertical displacement of core wall is 1413mm, appearing in the upstream of the core wall near the dam axis where the elevation is 657.00m at the deepest valley section. The maximum first principal stress of the core wall is 1839kPa, the maximum second principal stress is 1373kPa, and the maximum third principal stress is 1256kPa. Under normal water level condition, although core wall stress is smaller than depletion layer stress, but no tensile stress appeared, so the loam core wall won't produce pull crack. Meanwhile, for the strength and modulus of core wall is lower and the core wall is easy to adapt to deformation, core wall stress levels are lower. It indicates that the core wall is stable.

Cut-off wall
The maximum horizontal displacement of the first cut-off wall is 469mm, pointing to the downstream, and the vertical displacement is -112mm, and that of the second cut-off wall are 583mm and -129mm respectively. The maximum first principal stress of the first concrete cut-off wall is 9806kPa, the maximum third principal stress is 714kPa, and that of the second one are 16844kPa and 1094kPa respectively.

Finite element model
The finite element model what can be used to analyze earthquake responses of the dam is build, which is the same as the model for the static analysis.

Dynamic parameters and calculation condition
The time history analysis method is used and the equivalent nonlinear viscoelastic model is chosen as the constitutive model of soil, which assumes that the soil of dam body and foundation overburden are considered as viscoelasticity, and equivalent shear modulus G and equivalent damping ratio  are applied to reflect the nonlinearity and hysteretic nature this two properties of soil dynamic stress-strain relation. The dynamic parameters of dam are shown as in Table 4 and Table 5. Under the Wenchuan earthquake condition, in the dam site of Bikou hydropower station, the seismic intensity is comprehensively evaluated as degree 9. Correspondingly, the bedrock in the dam site has a peak level acceleration of 404cm/s 2 and its seismic acceleration curve of bedrock in horizontal direction is shown as Fig.7, which is in up-down stream direction and in dam axis direction. When calculating, the seismic acceleration value of vertical direction is regarded as two thirds of the horizontal one. Meanwhile, the dynamic water pressure is evaluated by the additional mass method called Westergaard method, that using equivalent additional mass instead of the dynamic water pressure to stack with the quality of the dam body during the earthquake.

Earthquake responses
Under the conditions of Wenchuan earthquake intensity, the earthquake response of the dam are analyzed, including the maximum acceleration response, displacement response and stress response of dam body, loam core and cut-off walls, and the earthquake induced permanent deformation (Serff et al., 1976;Taniguchi et al., 1983;Kuwano & Ishihara, 1988;Cascone & Rampello, 2003;Elia et al, 2011) of the dam is also obtained. The maximum values of earthquake responses analysis results are shown as in Table 8.The distributions of the dynamic analysis main features (Idriss et al., 19773;Ahmed-Waeil et al., 1990a, 1993bZhu et al., 2003) are shown as in Fig.8~Fig.14.

Acceleration response
Calculation results show that the natural vibration period of Bikou dam is about 0.74 seconds. The maximum acceleration response of the rockfill respectively is 4.08m/s 2 along the river direction, 4.43m/s 2 along the dam axis direction and 3.20m/s 2 along the elevation direction. For the core, the maximum acceleration response along the above three directions are 3.89m/s 2 ,4.14m/s 2 and 3.14m/s 2 respectively. Along the dam axis direction, the acceleration response of the dam is the greatest, followed by the acceleration response along the river direction and the acceleration response along the elevation direction is the smallest.
When earthquake comes, the upstream and downstream dam shell materials, which have a relatively long distance to the cut-off wall, are slightly affected. The corresponding horizontal acceleration response is basically proportional to the dam height and the maximum one appears near the upstream and the downstream slope surfaces which are close to the dam crest. The seismic acceleration response of a point is related to the vertical distance between the point and the bottom of the river valley; the height is greater, the corresponding seismic response is greater. As the slide slope of the right bank is slower than the left bank, the maximum horizontal acceleration response appears much closer to the right bank. From the middle of river bed to the both banks, the horizontal acceleration response decrease gradually which is related to the dam body type and boundary conditions. From the distribution of acceleration response on cross section, it can be seen that near the right bank, as the base of the downstream dam body is higher than the upstream and overburden layer exist at the bottom of the upstream dam body, the acceleration response of the upstream dam body is greater than the downstream dam body. The situation is opposite near the left bank, as deep overburden layer exists at the bottom of the downstream dam body and the downstream dam slope is much steeper than the upstream, the acceleration response of the downstream dam body is greater than the upstream dam body.

Displacement response
The maximum displacement response of the rock-fill is 25.92mm along the river direction, 16.99mm along the dam axis direction and 12.39mm along the elevation direction. For the loam core, the maximum acceleration response along the above three directions are 25.61mm, 16.77mm and 12.33mm respectively.
The displacement responses are not so great. Along the river direction, the displacement response of the dam is the greatest, followed by the displacement response along the dam axis direction. The smallest displacement response is along the elevation direction. The distribution of displacement response is almost the same with acceleration response. Under the seismic action, the displacement response is basically proportional to the dam height and the maximum one appears near the upstream and the downstream slope surfaces which are close to the dam crest. The seismic acceleration response of a point is related to the vertical distance between the point and the bottom of the river valley, the height is greater, corresponding seismic response is greater. The maximum horizontal displacement response appears near the dam crest above the deepest site of river valley. As overburden exists at the left bank, the maximum vertical displacement appears much closer to the left bank. From the center of river bed to the both banks, the displacement response decreases gradually which is related to the dam body type and boundary conditions. From the distribution of displacement response on cross section, the displacement response of the dam is not only related to the grade of upstream and downstream slope, but also the depth and distribution of overburden. Close to the right bank, as the base of the downstream dam body is higher than the upstream and overburden layer exists at the bottom of the upstream dam body, the displacement response of the upstream dam body is greater than the downstream dam body. The situation is opposite near the left bank. As deep overburden layer exists at the bottom of the downstream dam body and the downstream dam slope is much steeper than the upstream, the displacement response of the downstream dam body is greater than the upstream dam body.

Stress response
For the rockfill of dam, the maximum first principal stress is 425kPa, the maximum second principal stress is 352kPa and the maximum third principal stress is 203kPa. For the core, the maximum first principal stress is 265kPa, the dynamic tensile stress is 221kPa; the maximum second principal stress is 215kPa, the dynamic tensile stress is 178kPa; the maximum third principal stress is 192kPa, the dynamic tensile stress is 157kPa. For the first concrete cut-off wall, the maximum first principal stress is 3031kPa, the maximum second principal stress is compressive stress with a value of 591kPa and the maximum third principal stress is compressive stress with a value of 363kPa. For the second concrete cut-off wall, the maximum first principal stress is 4482kPa, the maximum second principal stress is compressive stress with a value of 683kPa and the maximum third principal stress is compressive stress with a value of 422kPa.
The stress response of rockfill and core along the river direction is the greatest, followed by the displacement response along the dam axis direction and the displacement response along the elevation direction is the smallest. The strong stress response of the core happens on the major river bed where the loam core connects to the both banks. The big and small principal stress response of the rock-fill become stronger with the increase of cover depth, thus the contour lines are parallel to the dam slope, and the maximum dynamic compressive stress and maximum dynamic tensile stress appear at the bottom of the dam near the deepest river valley. At the same elevation, the stress response of the dam body is much stronger than the core. As the great difference of deformation modulus between concrete and loam, stress concentration appears near the cut-off wall. Besides, at the downstream side of cut-off wall, dynamic stress response is very strong. From the distribution of stress response on longitudinal profile, it can be seen that the stress response at the bottom of the dam on the right bank is a bit stronger than on the left bank, the dynamic stress response is great at the connecting part between dam body and both bank slopes, especially near the place where the section of bank slope varies, stress concentration of strong dynamic stress response appears; and with the increase of peak acceleration, the stress response becomes stronger. So, more attentions should be paid to these weak areas.
Great dynamic stress appears at the top of the concrete cut-off wall, so does in the connection part between bedrock and the cut-off wall close to the bank slope. The stress response of concrete is proportional to the peak seismic acceleration, thus with the increase of peak acceleration, the dynamic tensile stress and compressive stress become greater. The maximum dynamic tensile stress of the cut-off wall is shown as in Table 6. Overall, the dynamic compressive stress of concrete is small and the dynamic tensile stress is relatively much greater, so the monitoring to the strength of the cut-off wall should be strengthened to ensure the safety operation of the dam.
During the earthquake, the maximum shear stresses of the dam are respectively 236kPa. On the cross section of the dam, the maximum dynamic shear stress and the dynamic shear stress become greater gradually from the upstream side and downstream side to the dam axis; but in the middle of the core, the dynamic shear stress decreases greatly, and at the bottom of dam, the dynamic shear stress response of rock-fill and transition material are very great. So, as the significant difference of the filling materials, the maximum shear stress responses are very significant in the connection parts of materials. The shear stress response of rock-fill is very great on the bank slope where the sections vary, and shear stress concentration appears in the local areas of where the sections vary at both banks; in the main river bed where the core connects to both bank, the shear stress is great.

Earthquake induced permanent deformation
The maximum permanent horizontal displacement is 74mm along the river direction, 47mm along the dam axis direction and the maximum permanent vertical displacement or settlement is 239mm. Taking no account of the thickness of the overburden on the dam foundation, the maximum dam height is 101.8m, then the permanent settlement induced by earthquake is about 0.23 percents of the dam height.
Due to the upstream water pressure, the earthquake induced permanent deformation of the core along the river orientation points to the downstream, so does the upstream dam shell materials near the dam crest. Along the dam axis direction, the earthquake induced permanent deformation is not so great and the maximum one appears near the upstream dam slope. It can be seen that in the dam abutments of both banks, the displacements of the dam body point to the center of the river valley; near the dam axis, the displacement and its variation gradient at the right bank are greater than the left bank. Affected by the concrete cut-off wall, the ultimate settlement of dam body below the elevation of 650m is small. And the settlement of the dam body above the river valley is basically uniform from the right side to the left, and the permanent deformation increases with the elevation increases. The maximum settlement appears near the downstream dam crest which is close to the dam axis. In general, the earthquake induced permanent deformations in both conditions are not great.

Earthquake-resistance safety evaluation 4.1 Comparing the static calculation results and monitoring data
In order to monitor the consolidation settlement in construction and operation, 5 collimating lines are set for the Bikou Hydropower Station soil core dam (Fig.18), which are respectively up dam 0-010.1m(708.00m elevation), under dam 0+007.8m(709m elevation), under dam 0+041.8m(691m elevation), under dam 0+093.0(670m elevation) and under dam 0+142.7(650m elevation). Only up dam section 0-010.0m got observed on December 17, 1975, and other sections started observation two years later after storage for the construction of crest parapet wall and downstream slope and drainage. Monitoring date of dam surface horizontal displacement and settlement before earthquake are shown as in Fig.16 and Fig.17.
Among all the monitoring sites, only section 0-010.0m started observation once storage, while other sections started observation relatively late. Therefore, displacement of these 5 sections cannot directly be compared with that of finite element calculation. As section 0-010.0m got monitored earlier, its monitoring date should be compared with modified results of finite element calculation, which can be shown as in Fig.16 and Fig.17.
According to the settlement distribution (Fig.18) along dam axis direction of section 0-010.0m, settlements of monitoring points on crest are approximately proportional to their corresponding rock-fill thickness at the same place. Crest position where locates on the deepest valley floor has the biggest settlement, and settlement on right bank is larger than that on left bank. Settlement distribution discipline of finite element calculation is roughly the same with that of monitoring results. Additionally, calculation values are larger than monitoring data, since dam settlement started monitoring later than corresponding calculation situation, then displacement before monitoring was included in the calculation results.

Results of earthquake-resistance analysis
In the process of dynamic calculation, safety factor development of every element, excluding bed rock, concrete cut-off walls, wave wall and joints, has been completely recorded throughout earthquake. The safety factor is defined as the ratio of shear strength to shear stress (including both static and dynamic shear stress) of element potential failure surface. As it to analyze the seismic safety of upstream and downstream, the anti-slide stability coefficient is calculated on pseudo-static method. Sweden Slice Method is adopted for slope stability analysis, where both horizontal and vertical seismic actions are considered.
Calculation condition for slope anti-slide stability can be seen in Table 7. The smallest safety factors of upstream and downstream slopes are shown as in Table 8, and the potential sliding surfaces are shown as in Fig.19. As the compacted earth dam design specification (SL274-2001) says that the smallest anti-slide stability factor under seismic action should be no less than 1.15. Under the condition BK-S1, slope stability of Bikou dam cannot meet the present criterion. However, on the conclusion of Bikou station injuring survey under "5.12" Wenchuan earthquake and its initial analysis report, only some joints connecting crest and body to embankment were damaged and needed amending, yet the whole dam was safe in general. Under the seismic action, safety factor of Bikou dam does not satisfy demand, however, the dam did not suffer from sliding or slope instability during the "5.12" Wenchuan earthquake. So, it is suggested that smaller safety factors should be adopted. For example, a level 2 dam has a safety factor ranging from 1.05 to 1.10. In a word, the safety factor distribution suggests that the Bikou dam has some partial scopes where the factors are less than 1 near the crest and upstream and downstream slopes. But the scales are sporadic and will not lead the failure of dam slopes.

Conclusions
In this paper, according to the actual engineering conditions of Bikou earth core rockfill dam, the 3-D non-linear FEM model of the dam is set up for calculating earthquake response by the dynamic time-history analysis method. By simulating the process of the filling of dam body and reservoir impounding, firstly the original static stress field of dam body is obtained under the normal water level. Then by the dynamic time-history analysis method, the earthquake responses of the dam, including acceleration response, displacement response, stress response and the earthquake induced permanent deformation of the dam are obtained for inputting the earthquake with peak acceleration of bedrock 404cm/s 2 . Under the action of the seismic with peak acceleration 404cm/s 2 , the acceleration response along the up-downstream direction is maximal with the value of 4.08m/s 2 , the one along the dam axis is secondary with the value of 4.43m/s 2 , and the one along vertical direction is the minimum with the value of 3.20m/s 2 . The maximum displacement response of dam in the three directions are 25.92mm, 16.99mm, and 12.39mm respectively, and the maximum principle stress responses of rockfill body are 425kPa, 352kPa, and 203kPa respectively. The maximum earthquake induced permanent settlement is 239mm, about 0.23 percents of the dam height. It is shown that the earthquake responses of the dam are close to the recorded data. The theories and methods for analyzing the earthquake responses of the earth dam here are feasible, and the results are consistence with the real situation of post-earthquake.