BBP Release Planning

Feature list for BBP Release v19.1

• Updated Irikura 1 (multi-segment, 500m GF)
• Updated GP (multi-segment, 500m GF)
• Updated Olsen (multi-segment, 500m GF)
• Update Seok Goo Song (multi-segment, 500m GF)
• Updated Exsim (500m GF)
• New Irikura 2 method (single segment, 500m GF)
• All methods support site response module
  • Phased out correction coefficients for Part A validation events
• New Validation events, with observations (Parkfield, San Simeon, Iwate, Chuetsu-Oki)
• GF for Central California
• Consistent approaches to implementing multi-segment methods.
  • GP and SDSU - multi-seg method 1 (one src file per segment, timing in SRC,timing in SRF file, merge seismograms)
  • SGS, IR1 - multi-seg method 2 (one src file per segment, Rupture generator produces single SRF file, output single set of 3 component seismograms)
• Added script to calculate number of sims needed for convergence
• Added calculation of all 3 components RotD100 and RotD50
• Migrated batch scripts to use Slurm jobs

Required for Release

• Part-A and Part-B validations at 500 m/s and get modeler approval
  • Irikura 1 - pending approval (Feb 4th)
  • Irikura 2 - pending approval (Feb 4th)
  • Song - pending approval (Feb 4th)
  • GP - approved by modeler (Dec 14th)
  • SDSU - under revision (Feb 4th)
  • Exsim - pending reply (Feb 4th)
  • UCSB - waiting 500m/s GF (Feb 4th)

• CSM
  • Remain on the BBP, but disable on user interface

• Update validation event packages
  • Remove amplification factors (Fabio)
  • Revise observations to take care or rotation issues, etc (Christine)
  • Update station list with low and high-pass limits (Christine)

• Irikura 2
  • Update license and documentation links

• SDSU
  • Code modifications (Kim, Rumi)

• Release evaluation
  • Update super table with new events (Fabio)

Changes Requiring More Time

• SDSU multi-segment changes (with updated GP method)
• UCSB multi-segment changes
• Updated Licensing statements in distribution
• Documentation on individual methods
• Migration to Python 3
• Development of CIG-level user documentation

Methods with Changes

1. Graves Pitarka - Multisegment (approved), 500m (approved)
   1. Pending - no
2. Olsen - Multisegment (approved), 500m (approved)
   1. Pending Review of method and new codes
3. Seok Goo Song - Multisegment (approved), 500m (in review)
   1. Pending - no
4. Irikura Recipe 1 - Multisegment (under review), 500m GF (approved)
   1. Pending - approval from AP
5. Irikura Recipe 2 - Single segment (approved), 500m GF (approved)
   1. Pending - no
6. UCSB - Multisegment (under review), 500m (in review)
   1. Pending - Multi-segment codes
7. Exsim - Single segment (approved), Parameter Changes 500m/s (in review)
   1. Pending - approval from Karin
8. Anderson - Single segment (in review, no 500m/s)
   1. Pending - no

Product Backlog

1. Shooting for a code freeze early January, followed by a release. Expected changes/work until then
   1. SDSU (Rumi, Kim and Fabio), implementation and testing of main code (perform hybrid merging in time domain instead of frequency domain)post-processors: inter-frequency correlation (almost ready), spatial correlation (may be possible)
   2. UCSB (Jorge and Fabio): complete the implementation of multi-fault rupture capability
   3. Irikura Method 2 (Hiroe and Fabio): ready, but must include the new disclaimer note in the code (sent to Fabio this week)
   4. GP (Rob and Fabio) complete tests for latest RG parameters, requires new Part B criteria for Vs30 = 500 m/s (attached) adjust high-frequency parameters
   5. Event files Christine and Fabio)
   6. Fabio to use new events files for Part A that include a Tmin limit (we only used Tmax in the past); some records need to be re-processed too. Christine to give all needed file to Fabio soon.
2. Christine generated new Part B criteria (low, mean, high) for Vs30=500 m/s, to be used with the new vmods (attached)
3. We agreed to not publish amplification factors with the event file anymore and people can either use the site effects module or have simulations at the vmod-specific Vs30.
4. We started to discuss the pros and cons of providing both sets of vmods (Vs30 of 863 and 500 m/s). Discussion to be continued.

Related Entries

BBP