[Bug 1073591] [NEW] Add one or more cases for SA IMT to hazard QA tests

1 view
Skip to first unread message

Lars Butler

unread,
Oct 31, 2012, 10:39:10 AM10/31/12
to oqb...@foldr3.com
Public bug reported:

Currently, all of the classical hazard QA tests use use a single IMT:
PGA. It would be useful to have 1 or more cases with multiple IMTs. SA
is a good example, since it has special handling in many parts of the
software for the SA period value.

** Affects: openquake
Importance: Medium
Assignee: Damiano Monelli (monelli)
Status: Confirmed

** Changed in: openquake
Status: New => Confirmed

** Changed in: openquake
Assignee: (unassigned) => Damiano Monelli (monelli)

** Changed in: openquake
Importance: Undecided => Medium

--
You received this bug notification because you are subscribed to
OpenQuake.
Matching subscriptions: openquake-bugs
https://bugs.launchpad.net/bugs/1073591

Title:
Add one or more cases for SA IMT to hazard QA tests

Status in OpenQuake:
Confirmed

Bug description:
Currently, all of the classical hazard QA tests use use a single IMT:
PGA. It would be useful to have 1 or more cases with multiple IMTs. SA
is a good example, since it has special handling in many parts of the
software for the SA period value.

To manage notifications about this bug go to:
https://bugs.launchpad.net/openquake/+bug/1073591/+subscriptions

Damiano Monelli

unread,
Jan 24, 2013, 8:59:35 AM1/24/13
to oqb...@foldr3.com
A QA test considering multiple IMTs can be constructed starting from
test 1 for classical PSHA (https://github.com/gem/oq-
engine/blob/master/qa_tests/hazard/classical/case_1/job.ini)

By setting:
intensity_measure_types_and_levels = {"PGA": [0.1, 0.4, 0.6], "SA(0.1)" : [0.1, 0.4, 0.6]}

The expected probabilities of exceedance for PGA must be the same of the
expected PGA values for test 1, while the expected values for SA are:

Probability of Exceedance SA[0.1] (0.1, 0.4, 0.6): 0.608675003748,
0.330831513139, 0.201472214825
Reply all
Reply to author
Forward
0 new messages