Minutes from 18 February 2009

2009/02/18 CSEP meeting

Attendees: Phil M., Danijel S., Jeremy Z., John Y., Masha L.

1. Status of CSEP V9.1.1 on csep-op:

   • SW Pacific one-year forecasts groups failed for 2009-01-14:
     
     1. John fixed permissions for directories, Masha re-processed test date successfully
   • John backed up old data, and freed up some disk space
   • John reported that new hard disks should arrive end of this week
2. Tasks for V9.4 release:
   • High priority:
     

     1. DONE: RELM S (space) and M (magnitude) evaluation tests

        • Available on csep-cert for RELM 5-year forecasts groups

     2. IN PROGRESS: Staging of existing raw catalog data to be used by re-processing
        • Code changes are in SVN repository, working on acceptance tests

        • Masha discovered lost reproducibility for first 3 operational months of raw one-day forecasts files due to:

          1. V1.0 didn't create metadata files for forecasts files as they were generated by the model codes
          2. V1.0 did create metadata files for forecasts files converted to XML format, but these files were re-generated due to discovered bug (Trac ticket #91)
          3. Solution is to manually re-store metadata files for forecasts:
             • It's very easy to isolate run-time Dispatcher directories that store input catalogs used to generate these forecasts
        • Everybody agreed to discard existing time-dependent forecasts and test results once they are re-generated with fixed ANSS catalog
     3. New web representation of test results:
        • Danijel suggested to create plots for all results on the web-server site
        • Acceptance tests for web viewing code
        • Will need to publish already existing test results in XML format, but extra care must be taken in posting re-processed test dates - only latest results should be posted
     4. IN PROGRESS: New web-page for the CSEP:
        • John and Danijel are changing colors for chosen theme
     5. Distribution of CSEP and dependent packages to other testing centers
   • Low priority tasks:
     
     1. Modular functionality of the CSEP Toolkit:
        
        1. Most of the modules can be invoked in standalone mode
        2. There is a lack of documentation on how to invoke modules:
           • Use Doxygen?
     2. New models:
        • Global testing region:
          

          1. Danijel suggested to install CMT version of Yan's global forecast model using CMT catalog (M>=5.8)

          2. Masha has sent email to Yan with model requirements for the submission
          3. Masha downloaded Yan's Fortran code to evaluate how easy it's to install within CSEP Testing Center
             • Should be the same amount of work to install the model as previously installed Yan's models for other testing regions:
               
               1. Get rid of hard-coded paths
               2. Use provided converter to generate binary format of input CMT catalog data
               3. Add support for input parameter file (forecast start date and time, location of input catalog and output forecast files)
               4. Write forecast converter to XML format
        • California:
          
          1. Masha sent email to Max asking if he will install ETAS model for V8.4.0 - Max is planning for V8.7.0
     3. Trac ticket #133: Allow specification of uncertainties to create modified catalogs
     4. Trac ticket #134: Capture timestamp within name for the log file
   • Re-processing due to Trac ticket fixes:
     
     1. #110: RELM Mainshock forecasts groups should use declustered catalog for evaluation tests
     2. #128: N-test delta should be computed as poisscdf(N, lambda)
     3. #114: Cutting catalog to geographical region eliminates valid events if their longitude or latitude values fall on the area cell borders
        • Should stage existing raw catalog, re-create observation catalog, and time-dependent forecasts
     4. #113: ANSS events with missing depth value have their time set to 00:00:00 by Matlab import routine
     5. #121: r-test results flipped
   • New hardware and OS upgrade to Fedora Core 10:
     
     1. John has new server set up, will become 'csep-cert' some time next week
3. GEM requirements for CSEP:
   • Should have a discussion after Phil gets back from GEM1 meeting
4. Statistical Seismology workshop hosted by SCEC in April