Minutes from 2 April 2008

2008/04/02 CSEP meeting

Attendees: Phil, Yan, Fabian, John, Jeremy, Amit, Masha

1. Introducing Western Pacific testing region into the CSEP:

   • Amit identified 2 extra events with Mw >= 5.8 in CSEP input catalog vs. UCLA input catalog:

     1. Yan updated his input catalog with one of missing events
   • Amit re-generated forecasts using Yan's input catalog and CSEP automatically generated catalog:
     
     1. Yan said that differences are mostly due to one event still missing in his input catalog
     2. Yan said that we should consider CSEP generated forecast converged with Yan's generated forecast
   • Amit is working on:
     
     1. Acceptance test for CMT catalog pre-processing
     2. Python class to automate Yan's forecasts within CSEP:
        • Create parameter file for the model with location for input catalog and result forecast file
        • Invoke the model
2. CSEP Version 8.4.0:
   • Installed on csep-op on Tuesday, April 1, 2008
   • Had to remove true cumulative results for one-day models N-tests:
     
     1. New implementation of the test uses different format
     2. Will need to re-process August 1, 2007 - October 27, 2007 and December 4, 2007 - March 1, 2008
   • John presented web-page for CSEP results
3. Software development:

   • Fabian is working on QuakePy re-factoring and acceptance tests

   • Masha is working on BatchProcessing module to automate re-processing of test dates in batch mode

   • Phil suggested to define a roadmap for the next Version 8.7.0 release in next week meeting.
     Tasks to consider are:

     1. Batch processing module
     2. Introduce Yan's models for Western-Pacific testing region
     3. Uncertainties for cumulative results
     4. Combine intermediate and true cumulative results in one plot
     5. Adopt QuakeML format for catalog data
4. Phil gave an overview of other SCEC projects that might have some impact on CSEP