Buried Peat problem

[eric1037]

Ok, I have a lulu of a project. Here's the scenario:

My client is planning on constructing three single-story, slab-on-grade retail structures on a site that has been filled periodically over the past 10 to 12 years. Column loads are in the range of about 80 to 100 kips. One of the structures is a long and skinny strip mall structure and the two others are chain restaurant-type structures. Our client wants to raise grades about 2 to 3 feet.

The soil profile generally consists of 23 to 26.5 feet of fill over 2.5 to 4.5 feet of amorphous peat and/or organic silt and clay. The PT/OL is underlain by clay with shear strengths in the range of about 2 ksf to about 32 feet, then sand with N-values of 25 to 40 to about 50 ft., then sandy silt with N-values in the range of 80 to 90 to 60 ft., then sand with N-values in the range of 80 to 90 to the explored depth of 70 feet.

Groundwater is below the PT/OL at a depth of about 30 feet.

About 18 feet of the fill was placed about 8 years ago. We have spotty compaction tests on the very top that show at least part of the final lift was compacted to 95% of a modified proctor.

N-values in the fill range from about 10 to 58, so it's not in that bad of a condition. However, we did encounter buried debris like pieces of concrete, etc. in the fill. I was also informed that a building previously occupied part of the site and they left an exterior concrete slab and some asphalt pavement in place.

The peat has moisture contents ranging from about 40 to 60 percent with N-values in the range of 16 to 30. Not too shabby. Based on this data, I believe the peat is currently in secondary consolidation.

Based on Schmertmann's method, the stress increase becomes negligible at 2 time the fdn diameter for column footings and 4 times the fdn width for strip footings. Therefore, the stress increase at a depth of 23 feet on the PT/OL should be fairly low, but I know that it doesn't take much stress incress on organic soils to induce settlement.

I am very leary about placing shallow spread footings on the fill, especially with an additional 2 to 3 feet of fill on the site. I haven't gone through the effort of doing a Boussinesq stress analysis yet, but it's my feeling that there will be differential settlement.

I considered auger-cast piles, but I think the downdrag of 23 ft. of sand, gravel, and debris would make the piles very long. Also, the debris and buried slab may make a nightmare out of pile installation.

I have considered double-cased drilled piers with permanent casing, but that just seems like overkill for structures like this.

I have also considered a raft or mat foundation, but one of the structures is a strip mall that is fairly long and narrow and I'm not sure how much it would help. I have asked the architect and structural engineer to consider seperating the individual units to help with differential settlement.

My last consideration is pre-loading the site for an extended period of time, say 6 to 12 months. However, the soil piles would be rather large and it may not fit into my client's schedule. It would present an unknown because we would have to install some instrumentation and monitor the settlement until we are satisified we are well into secondary consolidation. In addition, we would probably have to stage the loading and it could take much longer than anticipated.

There is a ray of hope in that the strip mall building area has not been completely filled and one half of the area is still at the top of peat level. We could potentially excavate the fill out of the area of the strip mall and use it to preload the other building sites. The organics could then be removed and the resulting excavation filled with engieered fill.

Ok, now that you know the situation I would like to know if I am missing anything or if I am being overly cautious. I am of the mindset that building over PT/OL should be done VERY cautiously with no guarantees because of the long term "tertiary" settlement that results from biologic processes and the potential for large differential settlement due to varying thicknesses and physical properties.

Any input?

Thanks in advance!

 

[GeoPaveTraffic]

I think you have a well considered opinion of the site and problem and are not being overly cautious.

I would lean toward preloading the building areas. I would look at placing enough preload to take care of the 2 to 3 feet of fill and forget about the building loads. The materials that will be stressed by the footings appear to be pretty good. If you could get the primary settlement out due to the additional fill the footing will be minor.

Another thing to consider would be to use as high a bearing pressure on the footings, bearing on the new fill, as possible. This will reduce the depth of influence and may make the differential settlement of the building that much less. It may even be worthwhile to overexcavate the upper part of the fill below the footings before placing the preload so the footing pressure can be increased.

 

[eric1037]

I like the idea of overexcavating the upper part of the fill. We could replace the sand with a gravel material to attain a higher bearing pressure. I have also considered using a system that Tensar touts called "Dimension", which essentially is a layered system of gravel and geogrid.

Also, the use of lightweight fill has crossed my mind.

These options are OK, but they may require more expertise on the part of the contractor. I still like the good old-fashioned pre-loading. It leaves the thinking to the engineers.

 

[DRC1]

I think a lot depends on the quality of the peat. Some peats are hih in organics, have high void ratios and will continue to decay and compress undertime. Others are more silty with low wer amounts of organics and are more like poor soil. Since the thickness is only about 30 feet, I would relook at piles, possibbly taper piles like Monotube that cann fetch up quickly in the dense sands below the peats.

 

[BigHarvey]

Your main problem is the settlement induced by the fill. Did you think about compaction grouting. If your peat is a real peat ( and not some kind of clayey silty sandy stuff ) it should react pretty well to compaction ( particularly when you consider that it is above ground water table ). Preload is also a good option since settlement is obtained quickly in peat. The choice between the two is governed by cost effectiveness.

 

[dmoler]

Really high N values for true worrisome organics.

I dont think that Schmertmann's method is really good for this type of thing.

I would suggest preloading as some have done above, but the verification would be problematic. I install settlement plates for shallow settlements, and after preloading, when the settlement is constant, call it good. You would have to use deep tube settlement procedures for this, which I have had problems with.

You could consider a structural solution, tell them to beef-up the reinforcement.

 

[eric1037]

I appreciate the responses. I am going to recommend preloading the site, beefing up reinforcement and doing a partial remove and replace to make the soils more consistent about 10 feet below the bottom of footing.

The more I look at it, the less I like a deep foundation alternative. There is too much downdrag, so the piles would have to be relatively deep. Besides there is buried debris, slabs, boulders, etc. that would have to be excavated to get through with any type of deep foundation system.

The problem I am having is providing an estimate of time for preloading. The more I look at the organic soils, the more I realize that some of it is more like buried topsoil with moistures in the 25 to 30 percent range, while some of it is more of a true peat with moistures in the range of 60 to 68 percent. The thickness is pretty variable as well. I am having a really difficult time estimating the properties without performing actual consolidation testing. Even if I did consolidation testing, I would have to do 5 or 6 samples and even then I wouldn't be sure that it's representative.

I have been using correlations from The Muskeg Engineering Handbook by Ivan C. McFarlin, but the estimates are for Canadian Muskeg peat, not the exact peat I have at this site. I am very conservatively estimating that the consolidation will take up to 4 months, which fits well with the construction schedule.

The other problem is monitoring the settlement to verify the peat has reached secondary consolidation. I have considered settlement plates, but they may not accurately record the settlement of peat 23 feet below the fill. Does anyone have experience with the magnetic extensometers for settlement?

 

[BigHarvey]

Water content of 60% is not peat but a clay with organic material in it ( real peat would show much higher walues for w ). Preloading should work but if you have organic matters you should have significant secondary consolidation ( creep ) which should be taken into account to calculate your surcharge height. I would consider as a first approach Cv value of 5 10-8 m2/s (17 ft2/year )to calculate consolidation time.
magnetic settlement gauge work fine but you only need them if you want to identify from which layers your settlements come from.
i would suggest that you simply use Asaoka's graphical method to monitor your settlements . You will be able to predict exactly time and amplitude of total settlements to come after a few weeks of monitoring.

 

[eric1037]

I think the 60% water content is due to the fact that the peat has been under 18 to 25 feet of fill since 1997. The unconsolidated peats in the areas that hadn't been filled were about 120%.

 

[GeoPaveTraffic]

I have used the "spider magnate" magnet extensometer system produced by slope indicator company. They are installed on inclinometer casing and read with a simple magnetic switch on a cable. I was using them in a very soft clay overlain with about 15 feet of medium stiff clay. They worked very well. Installation is tricky and requires a good driller and some extra time.

 

[BigH]

The thing that strikes me is that at one end of the proposed strip mall, there is no fill yet? That means you have a slope height of about 20 ft at this end - or is the peat/organic layer 'dipping'?
For the case of the single story chain restuarants and the strip mall on substantial fill, you have deep fill thickness over the worrisome zone. While the consistency in density and materials of the fill are somewhat suspect, the fact that other buildings have been there in the past is encouraging. Why does the client want to raise grade 2 or 3 ft? While theoretically 2 to 3 ft of fill can cause a 230psf increase of stress (areal fill) in the worrisome zone, there will have been some quasi-preconsolidation and I doubt that the fill would really cause much settlement - maybe less than 25mm. (We had very soft clays at a site and had to carry out stage loading - the first couple of metres caused almost no settlement (10mm max) whereas between 2 and 4 metres of fill caused 75mm to occur.) As for your building foundations, the strip mall should make the use of a grillage raft foundation a good choice - no need for full depth raft, then. The loadings should be small enough that, again, by the time the loadings are to 23 ft or so, very little real load is transferred to the worrisome material. At this time, you might wish to subexcavate a metre of so strip beneath the grillage foundation strips to ensure uniformity of the fill and density in the supporting upper metre.
I am more concerned, though, in the in the area of where you have peat/organic at the surface - you do not have the 20 some ft of fill. Here, if you are maintaining grade, you will require substantial fill (and presume that you have fill on a sloped "embankment end"). This may cause major settlements that are also uneven in nature (differential). You could, as indicated, dig out the peat/organic in the exposed areas, but you won't be able to underneath the sloped fill. This is the area where settlements may be a problem and you might need to extend down below the worrisome material. I would concentrate on the solution in this area.
BigHarvey is correct about the monitoring of preload if that is the way you go. While I am not sure you really need to, you might consider preloading the bulk of the area to ensure building out settlements and then using the preload fill to fill up the exposed organic zone, to replace the organic zone and to fill up the "embankment slope."