Site Class Assignment

[JLNJ]

ASCE dictates that we assign site class basd on the profile in the top 100' of soil. If there is a significant basement under the building, is the correct procedure to evaluate the 100' of soil below the basement level?

I have seen this done but I don't see it specifically addressed in the code. I can understand arguments both ways, depending on whether one believes the soil-structure interaction to be more global or local.

If one has a 20' deep basement and the top 20' of soil is poor, the inclusion of the soil (or not) in the determination of site class can have a very significant effect on the seismic design category.

Similarly, do ground improvement methods (e.g. GeoPiers or undercut and replacement or soil) improve the site class?

 

[RobertHale]

The NEHRP Provisions commentary (FEMA 450 or FEMA P-750) brush up against this subject. My reading of the commentary suggests that the investigation be done based on the first 100 feet of soil from the finished grade and then you can effectively discount the portion above the basement if you determine the soils "contribute very little to the response of the structure" (FEMA 450-2, pp. 30 or FEMA P-75, pp. 231). As for ground improvements if you undercut and replaced crummy soil, I don't see why you would not improve the site class so long as all the weak soils were removed. I am not sure how much improvement you could get from aggregate piers. I suppose if you got enough compaction and densification it would get better, but that if only a prima facie analysis. Talking to GeoPier or Hayward Baker would be the most productive regarding those improvements. I hope this helps.