KeyError: 'AtkinsonBoore2003Slab'

[Alessandro Bonazzi]

Dear members,

 

I've just started using the GEM engine on the seismic hazard maps released by the SHARE consortium (http://www.efehr.org:8080/jetspeed/portal/hazard.psml).

 

This data set comes with all the input files required by GEM and it seems to work fine except for a number of calculation sites.

 

For instance if I try to compute the hazard for point:(15.1E, 37N) I get the error below.

 

My site is somewhere in Sicily so it seems reasonable for the engine to compute ground motion equations also using the branch Subduction Inslab, however the uncertainty model (Atkinson and Boore 2003) seems to fail. I get the same error also with Lin and Lee 2008, Youngs et al 1997 , and Zhao et Al 2006.

 

Has anyone experienced a similar error?

Thank you very much,

Alessanro Bonazzi

 

openquake@ubuntu-32:~/SHARE/SHARE_OQ_input_20140807/SHARE_OQ_input_20140807$ openquake --run-hazard computational_settings.ini
[2014-11-21 15:42:42,287 hazard #16 - PROGRESS MainProcess/31934] **  pre_executing (hazard)
[2014-11-21 15:42:42,296 hazard #16 - PROGRESS MainProcess/31934] **  initializing sites
[2014-11-21 15:42:42,314 hazard #16 - PROGRESS MainProcess/31934] **  initializing sources
[2014-11-21 15:42:44,676 hazard #16 - PROGRESS MainProcess/31934] **  executing (hazard)
[2014-11-21 15:42:44,691 hazard #16 - WARNING MainProcess/31934] Could not find sources close to the sites in area_source_model.xml (maximum_distance=200.0 km)
[2014-11-21 15:42:44,731 hazard #16 - WARNING MainProcess/31934] Could not find sources close to the sites in area_source_model.xml (maximum_distance=200.0 km)
[2014-11-21 15:42:44,794 hazard #16 - PROGRESS MainProcess/31934] **  Generated 1 block(s) for ('b1',), TRT=Volcanic
[2014-11-21 15:42:44,999 hazard #16 - WARNING MainProcess/31934] Could not find sources close to the sites in area_source_model.xml (maximum_distance=200.0 km)
[2014-11-21 15:42:47,794 hazard #16 - PROGRESS MainProcess/31934] **  Generated 14 block(s) for ('b1',), TRT=Subduction Inslab
[2014-11-21 15:42:49,510 hazard #16 - PROGRESS MainProcess/31934] **  Generated 3 block(s) for ('b1',), TRT=Stable Shallow Crust
[2014-11-21 15:43:02,217 hazard #16 - PROGRESS MainProcess/31934] **  Generated 12 block(s) for ('b1',), TRT=Active Shallow Crust
[2014-11-21 15:43:02,623 hazard #16 - PROGRESS MainProcess/31934] **  spawned 30 tasks of kind compute_hazard_curves
[2014-11-21 15:43:02,853 hazard #16 - PROGRESS MainProcess/31934] **  compute_hazard_curves   3%
[2014-11-21 15:43:03,125 hazard #16 - WARNING MainProcess/31934] Revoking 1 tasks
Traceback (most recent call last):
  File "/usr/bin/openquake", line 9, in <module>
    load_entry_point('openquake.engine==1.0.0', 'console_scripts', 'openquake')()
  File "/usr/lib/python2.7/dist-packages/openquake/engine/bin/oqscript.py", line 455, in main
    log_file, args.exports)
  File "/usr/lib/python2.7/dist-packages/openquake/engine/engine.py", line 458, in run_job
    job, log_level, log_file, exports, 'hazard' if hazard else 'risk')
  File "/usr/lib/python2.7/dist-packages/openquake/engine/engine.py", line 254, in run_calc
    _do_run_calc(calculator, exports, job_type)
  File "/usr/lib/python2.7/dist-packages/openquake/engine/engine.py", line 296, in _do_run_calc
    calc.execute()
  File "/usr/lib/python2.7/dist-packages/openquake/engine/performance.py", line 48, in newmeth
    return method(self, *args)
  File "/usr/lib/python2.7/dist-packages/openquake/engine/calculators/base.py", line 111, in execute
    self.task_completed)
  File "/usr/lib/python2.7/dist-packages/openquake/engine/calculators/base.py", line 85, in parallelize
    oqm.aggregate_results(lambda acc, val: task_completed(val), None)
  File "/usr/lib/python2.7/dist-packages/openquake/commonlib/parallel.py", line 249, in aggregate_results
    agg_result = self.aggregate_result_set(agg_and_percent, acc)
  File "/usr/lib/python2.7/dist-packages/openquake/engine/utils/tasks.py", line 85, in aggregate_result_set
    acc = agg(acc, result.unpickle())
  File "/usr/lib/python2.7/dist-packages/openquake/commonlib/parallel.py", line 241, in agg_and_percent
    raise RuntimeError(val)
RuntimeError:
  File "/usr/lib/python2.7/dist-packages/openquake/commonlib/parallel.py", line 89, in safely_call
    res = func(*args), None
  File "/usr/lib/python2.7/dist-packages/openquake/engine/utils/tasks.py", line 174, in wrapped
    return task_func(*args)
  File "/usr/lib/python2.7/dist-packages/openquake/engine/calculators/hazard/classical/core.py", line 210, in compute_hazard_curves
    gsims = trt_model.get_gsim_instances()
  File "/usr/lib/python2.7/dist-packages/openquake/engine/db/models.py", line 2389, in get_gsim_instances
    return [logictree.GSIM[gsim]() for gsim in self.gsims]
KeyError: 'AtkinsonBoore2003Slab'

[Graeme Weatherill]

Dear Alessandro,

It looks as though you are missing an extra "S" in the key. It should be "AtkinsonBoore2003SSlab". The convention for subduction GMPEs is to append "SInter" for subduction interface and "SSlab" for subduction inslab.

The corresponding keys for GMPEs can be found in the docs (http://docs.openquake.org/oq-hazardlib), under the links for the corresponding GMPE (http://docs.openquake.org/oq-hazardlib/gsim/atkinson_boore_2003.html - in your case). The key will be the class name of each GMPE.

Hope this fixes the problem.

Thanks,

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.
For more options, visit https://groups.google.com/d/optout.

[Alessandro Bonazzi]

Thanks Graeme!

 

It works fine now.

 

I copy below the changes I've made to file 'complete_gmpe_logic_tree.xml' following your suggestion.

 

Best regards,

Alessandro

 

 

 

<!-- 5.0 Logic Tree for Subduction Inslab -->
        <logicTreeBranchingLevel branchingLevelID="bl5">
            <logicTreeBranchSet uncertaintyType="gmpeModel" branchSetID="bs1" applyToTectonicRegionType="Subduction Inslab">

                <logicTreeBranch branchID="b51"> <uncertaintyModel>AtkinsonBoore2003SSlab</uncertaintyModel>
                    <uncertaintyWeight>0.2</uncertaintyWeight>
                </logicTreeBranch>

                <logicTreeBranch branchID="b52"> <uncertaintyModel>LinLee2008SSlab</uncertaintyModel>
                    <uncertaintyWeight>0.2</uncertaintyWeight>
                </logicTreeBranch>

                <logicTreeBranch branchID="b53"> <uncertaintyModel>YoungsEtAl1997SSlab</uncertaintyModel>
                    <uncertaintyWeight>0.2</uncertaintyWeight>
                </logicTreeBranch>

                <logicTreeBranch branchID="b54"> <uncertaintyModel>ZhaoEtAl2006SSlab</uncertaintyModel>
                    <uncertaintyWeight>0.4</uncertaintyWeight>

                </logicTreeBranch>
            </logicTreeBranchSet>
        </logicTreeBranchingLevel>

On Friday, November 21, 2014 4:10:51 PM UTC+1, Graeme Weatherill wrote:

Dear Alessandro,

It looks as though you are missing an extra "S" in the key. It should be "AtkinsonBoore2003SSlab". The convention for subduction GMPEs is to append "SInter" for subduction interface and "SSlab" for subduction inslab.

The corresponding keys for GMPEs can be found in the docs (http://docs.openquake.org/oq-hazardlib), under the links for the corresponding GMPE (http://docs.openquake.org/oq-hazardlib/gsim/atkinson_boore_2003.html - in your case). The key will be the class name of each GMPE.

Hope this fixes the problem.

Thanks,

Graeme

On 11/21/14 4:03 PM, Alessandro Bonazzi wrote:

Dear members,

 

I've just started using the GEM engine on the seismic hazard maps released by the SHARE consortium (http://www.efehr.org:8080/jetspeed/portal/hazard.psml).

 

This data set comes with all the input files required by GEM and it seems to work fine except for a number of calculation sites.

 

For instance if I try to compute the hazard for point:(15.1E, 37N) I get the error below.

 

My site is somewhere in Sicily so it seems reasonable for the engine to compute ground motion equations also using the branch Subduction Inslab, however the uncertainty model (Atkinson and Boore 2003) seems to fail. I get the same error also with Lin and Lee 2008, Youngs et al 1997 , and Zhao et Al 2006.

 

Has anyone experienced a similar error?

Thank you very much,

Alessanro Bonazzi

 

open...@ubuntu-32:~/SHARE/SHARE_OQ_input_20140807/SHARE_OQ_input_20140807$ openquake --run-hazard computational_settings.ini