Reactivated settlements in peat

[jackson]

We have been asked to look at a 150 year old, 4m high, railway embankment constructed over approximately 5m of peat over chalk. Over the past 3 years the embankment has begun to settle (approimately 500mm of downward movemnt). The embankment was never a problem previously and I would assume that after 20 years or so most of the settlemet would hace occuredd and the peat would be pretty well consolidated. The renewed movemnets correspond with the recent creation of an area of wetland next to the site. Is there likely to be a connection? Also how can a rise in groundwater level in an already consolidated material cause renewed settlements? Any advise on how to analyse the problem would also be appreciated. There is no indication of any slope instability (other than some minor surface slips in the over steepened ballast shoulder)and there is no deflection of track side masts so I don't think that it is a slope failure.

Thanks

 

[KAM_imported]

Is it possible that creation of the wetland area caused a drop in the water table beneath the embankment? If so, the effective stress on the peat has increased, resulting in new load and settlement. Another posssibility is exposure of some of the peat due to a lowering of the water table, resulting in shrinkage of the exposed peat, and decomposition. Consider putting in groundwater piezometers to see what the water table is doing.

 

[jackson]

Many thanks for replying Kam. The water table now appears to be at ground level on both sides of the embankment. I am proposing to install piezomters within the embankment and also beyond the toe. I am also proposing to install some basic slip indicators to confirm that it is not a slope stability problem. Am I right in thinking that a rise in groundwater will lead to a decrease in effective stress and hence no settlement can occur?

 

[BillHolt]

Jackson,

I'd be real concerned about modeling peat as a clay. My experience is limited to testing a waste material that was very much like a peat, but the curves were not at all like what you'd expect to see with a clay. Rather than consolidation, compression predominated which largely rebounded on unloading.

I wonder, if is possible that you have a drainage layer of some sort at the peat/chalk interface? Perhaps unhealed fractures in the chalk? If either of these is the case, then your new wetland may, in addition to providing homes for mosquitos, have simply increased the load on the peat causing your added deflection.

Good luck, and please report back on this one as it's not a problem everyone will have experience with.

Bill Holt

 

[KAM_imported]

Peaty soils can create some pretty acidic conditions--is it possible that there has been some degradation of the chalk due to acidic groundwater below the embankment?

 

[jackson]

Thanks guys, your ideas are a great help. I'm doing an intrusive investigation next week and will inform you of the outcome. As it stands at the moment it seems to me unlikely that it is purely a settlement problem within the foundation even though all the observed movements seem to indicate that it is. Your degradtion of the chalk theory was intersesting Kam, as I was thinking today about the possibility of solution features at depth.

 

[tomsklucki]

Has the amount of rail traffic on the embankment increased in recent years?

 

[dimitrios]

Hello jackson,

although there have been already answers that cover beyond my knowledge possible causes I would just like to point out that there are perhaps one main cause that triggers some others. Since you asked about the decrease in effective stresses due to water level rise, you are right that since they get smaller there will be also smaller deformation. Never-the-less increase of pore pressure would lower the shear strength and in that case have deformations due to local failure. I don't know if the small slip indications can be integrated in any failure mode.

If I may I would like to agree with tomsklucki's question also for the possible change of the type of trains (heavy loads instead of/or increased frequency).

Please consider the above as sharing thoughts (probably already known) that are neither an expert's opinion nor necessarily correct.

Regards