Hi Shafi, if you have access to the original hazard curve of PGA for or near the site of your project, you can get easily the PGA for 300 years of lifetime. For example, 50 years lifetime with 10% of probability of exceedance yields about 500 years return period of ground motion (T=50/0.1) is the usual case, it  means that the number of earthquakes per year in this case is 1/500 = 0.002. Then if you have the hazard curve look for the number of earthquakes per year that matches 300 years life time with 10% of probability of exceedance (300/0.1=3,000 years return period) that is equivalent to 0.0003 earthquakes per year in the "y" axis. So just intercept the hazard curve to this value (0.00033) and get in "x" axis the PGA you are looking for. If you need the PGA for another probability of exceedance (i.e. 2% instead of 10%) you must divide all the former calculations by 0.02 instead of 0.1. If you do not have the hazard curve, you have to perform a hazard assessment for your site using available information. Hope it helps.

In some countries there is an engineering practice to double PGA when moving from smaller life period to greater one (e.g. double when moving from 100 years to 500, and double again from 500 to 1000). I think the similar rule is used in Pakistan: double PGA when transferring from 50 to 300.

Dear Sergey Kuznnetsov

Thanks for your response. I think it will work as a first guess. But it seems to be a thumb rule and more conservative. Could we also use statistics to calculate the PGA value for 300 years more accurately?

Dear Waltar Salazar

Thanks a lot for your detailed answer. You have very nicely (as well as simply) calculated the return period (T) from life (n) and probability of exceedance (Pe). I was using a bit complicated formula as follows:

Pe = 1-(1-1/T)^n

with Pe=10% and n=50 years, it gives T=475 years (instead of 500 years). Also 

with Pe=10% and n=300 years, it gives T=2850 years (instead of 3000 years). Could you please tell me which one (of the 2 methods) is more recommended?

Secondly, suppose, I do not have the seismic hazard curve for the site and want a preliminary guess the PGA value, then would you suggest a standard hazard curve of some European or American Codes which could be used for a preliminary estimate of PGA?

Hi Shafi, you are welcome. Your way of calculation the return period is better and accurate.

Yes, you can get some preliminary PGA values once you have in your project similar tectonic conditions and maximum magnitnudes from Europe of the States Codes. Check also IBC code (International Building Code) if they have some PGA values for another parts of the world.

Hope it helps

Walter

Dear Shafi,

The answers to your question depends on which approach, probabilistic or deterministic/scenario, is used. As demonstrated by Walter and you, the hazard curve derived from probabilistic seismic hazard analysis (PSHA) could give you the value for the 10% probability of exceedance in 300 years. You could get a much higher value if you consider a 2% probability of exceedance in 300 years. However, these probability calculations are simply numerical manipulation and no scientific basis (see attached paper). 

I would suggest that you consider a deterministic/scenario approach to determine the value given high seismicity in your country.    

Article A critique of probabilistic versus deterministic seismic haz...

Dear Zhenming Wang

Thanks for your response and paper. These are very helpful.

Actually I use have a deterministic study value of 0.2g PAG for 10% exceedance in 50 years. What will be the value for the 10% exceedance in 300 years? Can I use some statistical calculations / formula for this prediction? Because I do not have any seismic hazard curve. 

Deterministic analysis derives a ground motion value from the maximum credible (or considered) earthquakes. No so-called seismic hazard curve is derived from the deterministic analysis. Also, deterministic analysis doesn't provide a probability of exceedance for a given time period such as 50 or 300 years.  

Dear Zhenming Wang

Thanks for your response. It means that first of all, I shall have to go for Probabilistic Seismic Hazard Analysis (PSHA) and then decide about the probability of exceedance in a given design life.

How to calculate PGA by involving rock mechanics parameters such as Vp, Bulk modulus, shear modulus, poisson ratio etc.

Dear Jahangir, Rock mechanics or Geotechnical parameters help in knowing the local soil amplification factor. If you know, PGA at the base rock, you can get the PGA value by multiplying it with the local soil amplification factor.

Hope this helps.

HOW TO CALCULATE THE LOCAL SOIL AMPLIFICATION FACTOR FOR 300 YEARS?

Walter Salazar

Hi Walter Salazar, Hope you are doing well.

how to prepare a catalogue/source model for a region in order to carry out PSHA? i mean how to find out the following values for a source model like min and max magnitude values, the seismicity of the particular zone, the likelihood of small or large earthquakes in the region and the exceedance rate.