Minutes from 6 April 2010

2010/04/06 CSEP Meeting

Attendees: Jeremy Z., John Y., Masha L.

1. csep-op:
   1. CSEP V10.4.0 is installed as of Thursday, April 1, 2010:
      • Passed all unit/acceptance tests
      • Announcement is sent out to csep-all distribution list
      • Due to new functionality (Trac ticket ##178 Add a sanity check for number of observed events in cumulative result)
        1. #213 Inconsistent number of events for 2008-05-02, 2008-06-02 observation catalogs for CA RELMAftershock groups
        2. #214 Inconsistent number of events for 2008-05-03, 2008-06-03 observation catalogs for CA RELMMainshock groups
        3. Masha started re-processing for test dates of 2008/06-2008/10 that are affected by drop in number of events in raw ANSS catalog
   2. Performance issue due to new NFS server:
      • Mentioned above re-processing is very slow
      • John wants to try separating testing regions to different disks
        1. Masha said it's OK since all testing regions are independent from each other
2. Jeremy ran tests with 10000 simulations for Western Pacific one-year-models-V9.1 experiments in stand-alone mode:
   • Results are consistent with the ones Masha generated on csep-op with 1000 simulations
   • Masha asked about necessity of fix for Trac ticket #180 Stabilize evaluation tests that use simulations by increasing number of simulations
     • Jeremy responded that scientists interested in results will re-process with higher number of simulations anyway, so it's OK not to implement it within CSEP testing framework
3. From 2010/01/26 meeting: Jeremy had a discussion with Peter Shebalin:
   • Should run alarm-based evaluation tests for EAST (alarm-based) model against EEPAS and PPE David Rhoades models (rate-based)
     • Masha responded that EAST model has a single magnitude bin vs. rate-based models having multiple magnitude bins
     • Jeremy said that alarm-based tests won't work then
     • Jeremy will get back to Masha regarding alarm-based tests for alarm-based model vs. rate-based model
4. Software development:
   1. From 2010/03/02 meeting: Masha asked how up-to-date we should keep our software stack:
      • Thomas Goebel installed CSEP generic package with later version of matplotlib library, plots don't look right
      • Phil suggested a discussion within larger group
   2. Self-discovering web application to view CSEP results:
      • Ankur started working on mod_python application
   3. Tasks for CSEP V10.7.0:
      • Get rid of remaining Matlab codes (#68 Adopt QuakeML format for catalog data):
        1. QuakeML format is not adopted by ANSS and CMT authorized data sources to distribute the data
           • Masha started replacing Matlab codes with NumPy functionality to prepare catalogs
           • Will need to implement catalog uncertainties (QuakePy does not support), will propagate to QuakePy
           • Will switch to QuakePy once data sources adopt QuakeML
        2. #51 Consider rewriting/invoking Fortran declustering code in Python
           • Masha contacted Thomas van Stiphout at ETH since he has adopted ZMAP catalog format by Fortran declustering code
   4. Pending new models:
      1. Max may submit ETAS model for California some time in March-April 2010
      2. Matt promised Java version of STEP to be available for the installation within SCEC Testing Center by March, 2010
         • Matt will get back to Masha if he wants to continue testing of fixed Matlab version
   5. From 2010/02/02 meeting: High resolution Global testing region:
      1. Danijel reported that forecast template and format are being defined
      2. Yan sent out an announcement about "FULL EARTH HIGH-RESOLUTION (0.1x0.1 degree) EARTHQUAKE FORECASTS"
         • Using PDE catalog
         • Will integrate within CSEP when testing region is established
5. From 2009/12/08 meeting: CSEP policies
   1. Should discuss if SCEC Testing Center results should be open to the public
      • Danijel will get back to the group
   2. Danijel will work on more content for the main CSEP web pages
6. Hardware:
   1. csep-cert
      • CSEPWebTest.py is not part of nightly build - some issues connecting to the web server
        1. John is looking at the problem