Blunder Question on annual probability of exceedance

[Sujan Raj Adhikari]

Hello Openquake user group

if I want to calculate the annual probability of exceedance of 2% in 50 years,5% in 50years, 10%in 50 years, 20%in 50 years 40%in 50 years and 50%in 50 years

how can I define it in Hazard output? is following the correct way or? please give me your thoughts.

[Hazard output]

export_dir = /tmp

individual_curves=true

mean_hazard_curves = true

quantile_hazard_curves = 0.15, 0.50, 0.85

hazard_maps = true

uniform_hazard_spectra = true

poes = 0.02, 0.001, 0.005, 0.0021, 0.001, 0.000404

regards

Sujan

[Graeme Weatherill]

Dear Sujan,

Your numbers are partially right. The probabilities of exceedance given in the poes correspond to the investigation time that you set, which should be 1 in the case you show.

You could either set the investigation time to 50 years and then define the poes as 0.02, 0.05, 0.1, 0.2, 0.4 and 0.5 respectively, or set the investigation time to 1 year and set the poes to 0.000404, 0.001025, 0.002105, 0.004453, 0.0101645 and 0.013767 respectively. Remember the relationship between rates (lambda), probabilities of exceedance (P) and investigation time (T) are expressed through the Poisson formula: P(T) = 1 - exp(-lambda . T).

If you are using the one year investigation time and corresponding poes then try to avoid rounding too much, especially at the higher probabilities.

I hope that helps.

Regards,

Graeme

--
You received this message because you are subscribed to the Google Groups "OpenQuake Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openquake-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openquake-users/7b6f7a6b-d957-4317-805a-08a3e39dc89en%40googlegroups.com.

[Sujan Raj Adhikari]

Thankyou Graeme, thankyou verymuch for this answere.