Seismic Site Classification

[Steel1040]

The problem is I am a new graduate and the other problem is a new structure is going to be placed on a site with a Class D seismic site classification. After running through the ASCE 7 classification, using data from our soil boring and SPT testing, it is a Class D and therefore would incur several hundred thousand dollars worth of structural improvements. I have been researching sub-grade modifications using pressure injected grout, geo-piers, etc, but cannot locate a specific code or specification that addresses how to handle an initial soil failure, and re-classification of the corrected site. If anyone can direct me to a book or a document (preferably online) that would assist me or has any advice after having gone through a similar circumstance, I would definately appreciate it.

Thanks in advance

 

[molerat2210]

What's the big deal about site class D? Most of the world is site class D. Use seimic refraction to get a better site class if you want.

Why $ for improvement? Is this a seismic retrofit job? If the building was not designed to handle D then retrofit to handle D. Our job is to tell the truth even when it hurts.

Ground improvements are not used to improve site class. That would be a waste of money and difficult to prove the benefit. Ground improvements can be used to mitigate liquefaction for site class F.

 

[Steel1040]

Something the Architect and the SE came up with. Somehow/way they were able to break out the difference in additional steel/materials/labor and come up with an actual cost for the different site classifications. And no, this is a new multi-story structure to be constructed at some point in the near future.

At this point in my 3 week long career, I just reasearch what they want and deliver the facts, which I already told them that based on our soil borings, the site falls under the Class D classification when using both Blow-Counts and undrained shear criteria but would need to verify this with the lead Geo-tech when she returns next week. When using ASCE 7, eq 20.4-4, the only way it would pass would be if you could re-classify rock as a cohesive material, therefore my dc value would be 100' instead of 66'... unless of course you are supposed to use d_c = 100' - (all layers of non-cohesive material excluding the rock).

Basically I have a 72' boring depth after that point we hit a gypsum layer. From the boring log, we had 2 sand seams of 1' and 3' respectively, and a clayey-gravel seam of 2'. From my interpretation of the code, d_c = 72' - 1' - 3' - 2' = 66'. If have made a mistake, I would like to get this corrected before I present it to her.

 

[grabens]

The SPT method can be conservative. If it is border line between a Site Class C and Site Class D, it may be prudent to run a field test such as ReMi or Seismic CPT, etc. to measure shearwave response on the site. If the SPT method is not close to a Site Class C, you are going in the wrong direction.

 

[grabens]

I'll add, unless you have shallow dense to very dense soils or shallow rock, you will be hard pressed to get a site class better than "D".

 

[jgailla]

molerat has it right. If there are several hundred thousand dollars at stake, it may be worth it to do a seismic refraction study for a few thousand dollars to see if the site class can be upgraded.
You should be able to find an engineer who specializes in these studies. One is out of the University of Virginia, I believe, but it has been a while since I've had any dealings with this issue.
Google should be able to find something.
Of course, there is always the chance that the study will not increase the classification, but SPT is conservative in this area. My experience is that F class can often be upgraded to D, but D to C is less likely without the presence of dense soils or rock.
It also depends on how close you are. If you are a D marginal to F, you're probably stuck. If you're D marginal to C, the study may be worth it.

 

[Steel1040]

The biggest issue is the layers of clayey and silty sands (3' layer) that occurred right around the groundwater table. The soils directly above and below these layers had an average undrained shear strength of 3500 psf, however these seams displayed results that were slightly above 600 psf and skewed the results lowering the overall s_u to around 1500 psf (well above an F, but not enough for a C). If these layers were brought up to a level that were closer to the s_u of the layers immediately adjacent, everything would be wonderful, hence my initial question regarding sub-grade modifications to increase undrained shear.

In any regard, I have begun looking into the Seismic Refraction Study. I just want to give the client as much factual information as possible to allow them to make the best decision on what to do with there building.

Thanks again, I truly appreciate it.

 

[jgailla]

From what you've said in the last post, it looks like it could be worth it.
Just make sure the client knows he's taking a gamble.

 

[fattdad]

Site class is baaed on the average conditions in the upper 100 ft of the soil profile.

Seismic refraction can only ascertain "layers" when the refracted velocity increases with depth (i.e., you won't be able to "see" a lower velocity layer below a higher velocity layer).

The seismic cone can give you a profile that shows higher and lower velocity layers throughout the soil column. They're cheap and fast too! (No, I'm not an in-situ testing contractor!)

f-d

¡papá gordo ain’t no madre flaca!

 

[Steel1040]

I was reading about that particular type of study and am leaning towards it. We are located in the midwest (Colorado, Oklahoma, Texas, Kansas, Missouri, Nebraska) and am having some difficulty locating a firm that will provide this service. Does anyone know any that are located within this region?

One other question, just for the purposes of seeing if this study was warranted, I had located a table of Mean Shear Wave Velocities in a Steel Design Manual(Borcherdt 1994), and was wondering how accurate these are or if anyone has had experience using a soil boring to develop the profile/characteristic, then using this table to determine the shear wave velocity? Again, I stress that using this table was just a quick evaluation to determine if using the shear wave velocity would be beneficial... and to help any future geotechs that may run across this problem

 

[GeoPaveTraffic]

Steel1040,

I recommend that you have a ReMi survey performed. It is quick and based on what you have posted has a decent likelihood of rasing your site classification. Additionally, a site specific response specturm can be determined for the site. This would likely lead to a more effiecent design.

The firm I work for, Shannon & Wilson, and many others perform these services in the mid west.

Good luck.

Mike Lambert

 

[molerat2210]

GeoPave,

The reason ReMi gives you a higher site class is for the reason Fattdad said. Surface wave tests cannot see lower velocity layers under higher velocity layers. The site clss could be goverered by a thin layer of weak soil. Therefore, ReMi can often provide a bogus higher site class. The proper way would be with downhole tests. This seems to be the intent of ASCE 7 / IBC since they want a weigthed average velocity of each layer.

 

[moe333]

I could be wrong, but I believe ReMi can see lower velocity layers below higher ones. Granted, it is likely not as accurate as a downhole test, it would be acceptable for site class determination.

 

[Steel1040]

Spoke with the State Geologist, and he informed me that his "official" position was that the REMI testing yielded comparable results to that of the Crosshole. He did not state what the statistical variance was, but, during state sponsored trials, the two tests yielded similar results.

 

[GeoPaveTraffic]

molerat2210,

I agree that downhole is better, however, ReMi is an acceptable method. Done properly by an experienced geophysist, ReMi provides an average value over the upper 100 feet.

Cost of ReMi is usually $2,000 or less for a site, while downhole is upward of $6,000 and cross hole even higher. There is no "right" answer, all depends on the site, type of construction, costs assoicated with extra detailing for a lower site classification, etc.

Mike Lambert

 

[ncgeo]

IBC Site Class can often be improved from E to D or from D to C by using shear wave velocities, since the IBC VS30 method can incorporate the rock velocities in contrast to just using the N-values of the soil. I have performed numerous surface wave surveys for this purpose in the Southeastern US. The ReMi method is just one of several surface wave methods used to obtain shear wave velocity profiles. The other methods are SASW (spectral analysis of surface wave methods, Dr. Stokoe) and MASW (multi-channel analysis of surface waves, Kansas Geological Survey). All three methods will provide reliable results if performed properly by a qualified and experienced geophysicist. I would recommend using an active weight drop source such as a sledgehammer instead of relying on only background noise, which was the original approach of the ReMi method that may have provided unreliable results in some cases due to lack of background noise energy.

The other methods of measuring shear wave velocities include seismic refraction, crosshole seismic, downhole seismic, and seismic CPT. However, seismic refraction typically does not "go" deep enough for site class purposes, while crosshole and downhole methods are more expensive than the surface wave methods, especially if you include the cost of drilling and casing the borings. In soft coastal plain soils, seismic CPT can be comparable cost to a surface wave test and you also get the typical CPT data also.

The cost for a surface wave survey generally starts at about $1500 US and will increase from there in proportion to the mobilization distance.

 

[molerat2210]

The only reason the site class comes out higher using REMI or any other method based on measuring wave velocity from the ground surface is that the thin, weak layers are obscurred. These methods do not meet the intent of ASCE 7 where the seismic velocity of each layer should be determined. I see this done improperty very often by purported experts. If the goal is to save someone money and create a paper trail of legitmacy then go for it. The formula is not an average of the top 100 feet it's a weighted average where thin, weak layers have a disprortinate effect on the site class outcome.

 

[dgillette]

Have you tried Zonge(sp?) in Denver?

 

[Steel1040]

We contacted Zonge several months ago and recieved an estimate from them to do the job, we unfortunately had to defer back to the Structural group who left it in the hands of the Client. Since then I have not heard what happened and since we were not directly involved, I assume that the site classification was left as is, and nothing more came of it. I was really looking forward to working with the Zonge guys and learning first hand about the procedures and what not. If the need arises for a "more refined" site classifcation, at least we know who contact right away. Oh well, maybe on the next multi-story government job site.