Page 1 of 1

seeking for HVSR modelling codes

Posted: Sun Nov 29, 2009 11:25 am
by gdm
I'm trying to work for a commercial software for (among the other things) analyzing microtremors.
Is there anyone interested in implementing the part related to H/V modelling? (as input: Vs Vp thk density Qp Qs (Rayleigh+Love) number of modes to adopt); output: HVSR + single-component spectra)


Posted: Fri Dec 04, 2009 5:41 pm
by admin
For what purpose? To invert experimental HV curves to retrieve model parameters? Experience has shown us and various papers confirm it that HV curves are the result of many uncontrolled factors (source location, Love/Rayleigh partition, excitation of surface higher modes, body wave contributions,...).

Anyhow, if you have paper references and if you can propose a method, you are welcome to contribute. I can help for the code integration but I cannot write core modules around this topic (lack of time).

This is certainly something a little bit different from what I understood in your message, but we have a Sh, Psv transfer function computation written in Fortran that should be included soon.



HVSR modelling code (based on surface wave ellipticity)

Posted: Sat Dec 05, 2009 9:09 pm
by gdm
Hi Marc and thanks

I do certainly agree with your points: it is barely impossible to soundly retrive vertical Vs profile from H/V data (unfortunately it is plenty of people who pretend to do it).

As I'm working with both surface wave dispersion and H/V methods I know that it is quite hard to get a model that matches both things...

What I'm trying to do is "simply" a semi-quantitative joint modeling/inversion via Pareto front analysis (to implement in a commercial software I'm working on).
And for such a thing I'd need a tool based on ellipticity (not simply body waves - I already have a tool for that).

So in case you wanna cooperate for such a thing we can maybe continue the discussion via email.

thanks again