gmf data with custom_site_id for scenario risk calculation

[dyn...@gmail.com]

Hi,

I noticed that gmf data created for a scenario risk calculation consisted of custom_site_id, which is neither from the input exposure nor from site_model. Wondering the logic behind the custom_site_id, and where I can find it in the open quake code. Also wondering if the same logic is applied in classical and event-based risk calculation. 

Thanks.

Regards,

Hyeuk

[Peter Pažák]

Hi,

I can see in the code that if you e.g. do a tile specification of sites, the custom_site_id is created using geohash:

https://github.com/gem/oq-engine/blob/master/openquake/commonlib/readinput.py#L772

I guess this approach was just chosen as a solution to identify the sites uniquely.

Peter

Dátum: nedeľa 22. júna 2025, čas: 9:06:52 UTC+2, odosielateľ: dyn...@gmail.com