Designing core of a dam in seismically active zone

[georam]

For an earth core dam with outer granular shell layers, the designed for the core of the dam is usually based on economic and availability of the fill materials. If the impervious fill material (clay) is readily available and cheaper than granular fills, the dam can be designed with relatively wide core, say with core slopes of 1H:1V. A dam with wide core, however, sometimes has problems during construction due to excess pore pressure developement (stability analysis may reveal that it will not meet the minimum factor of safety for temporary conditions). In addition, placing impervious fill may take longer than placing granular fills, and therefore will slow down
the construction progress. In cold climate, impervious fill placement is also limited during the summer periods only.

On the other hand, if the clay materials has very low permeability, the core can be designed to be smaller in size. Slopes can be steepened to even 1H:4V as long as no excessive seepage or high hydraulic gradient occurs along the downstream slope of the core. This would result in a dam
with almost vertical clay core and a large volume of outer granular fill zones.

My questions is the following :
If this dam would be located in a seismic active zone (say 0.15g or 0.2 g), are there any papers/technical documents which mentioned or supported that a relatively steep clay core should be avoided ? Or is it the other way around, ie. a seismically prone dam should not have a core
with flat slopes, say 1H:1V.

Imagine this, if you have an almost vertical clay core with outer granular fill dam, and it is shaken during earthquake, would the clay and the shells separated or an almost vertical crack be developed ? What happened if the clay core has 1H:1V slope. Will the contact with the outer
granular shells be better during shaking ?

Any comments or anyone can put some light on this one?

 

[Dragond2003]

A central clay core in a rockfill dam is usually designed to achieve a hydraulic gradient of 2 over the height of the embankment unless there is a specific need to reduce this (shortage of clay material, sloping core construction). The moisture content of the clay would also normally be about 2% wet of optimum and be well compacted to produce a relatively plastic material (in terms of behaviour that is).
The previous shells upstream and downstream will support the core with no problems of stability as a result of increased pore water pressure in the core. Slip failures in cores of rockfill (gravel) fill dams as a result of pore water pressures are almost unknown.

With regard to placing the materials, there is no need to place the different fills to maintain a level surface across the dam. Differential elevations in the shells and core are permissible as long as safe working conditions are maintained. The core should be supported by the filters and sufficient transition to provide a working track for access and differential levels controlled by considerations of temporary slope stability. In wet waether it would be normal to place the granualr fill and in the dry the clay core. This will minimise programmem time.

Regarding seismic activity. No well compacted fill dam has failed as a result of shaking to the best of my knowledge of 30+ years in dam engineering. The only likely result if compaction achieves more than 90% is some permanent deformation (settlement). Separation of the core and shells is highly unlikely.