1. Filter median empirical aggregations
2. Stampede2  
3. Double check Vs30 perturbation (James for Sarah)
   
   

James

1. Deferred
2. Robin taught how to use. Sung debugging
3. Done, a table of vs30 medians and sigmas is passed to the perturbation file

Claudio/Jason/Sung

1. Integrate HF 5.4.6 into the workflow (test data collection)
2. HF Rayset default off
3. GeoNET code Python3-ize
4. Include HF 5.4.6 into new HF/CUDA
5. Workflow change: HF to run separately

James, Viktor

1. WIP: Collecting test data currently. HF version 6.0.3 has just been received too
2. .
3. .
4. Complete, tests pass

HF CUDA Conversion

automated HF workflow on Hypocentre

1) UHS plots for different number of periods (background)

2) Fix epicentre benchmark tests (background)

2.5) Fix up context map and vs30 map.

3) seistech frontend todo:

• Enable lat/lon location selection
• Disagg
  • Epsilon plot
  • Download of contribution values
  • Download of mean values
• Integrate Context map (once done by Viktor)
• Integrate vs30 map (once done by Viktor)

4)Production wiki page

5) Ensemble wiki page

6) vs30  verification

7) Generate data for 22 sites (requires new IMDB creation - Jason)

8) Ensemble functionality for 12 sites: produce (figure and csv file) different branches of the ensemble+ensemble mean (currently only flt+bck = total plots/csv, not branch comparison)

9) GM selection linking Daniel's work

10) Performance enhancement (Empirical MPI)

11) DB Creation for vs30 check / NZGS

12) Distributed Seismicity magnitude distance scaling

Claudio, Jason, Daniel

1) No progress

2) Waiting on PR

2.5) Changes completed.

3) Done + started on additional seistech tasks

4) Done (Production - TODO)

5) Done (Ensemble Versions)

6) No progress

7) Done, also updated SHA script to use GMT disagg

10. GMT plotting works for epsilon and type using seistech data (to_json) except -0.5, 0.5 epsilon boundaries (missing)

11) DBs for vs30 test and NZGS partially created.

12) Implemented with a custom distance. 0km is off, default is 2 km.

13) Implemented

Roadmap / TeamGantt

1. Addition of new IMs from the list
   
   1. Fourier Sp
   2. SA _ (x,y,GM, RotD50, RotD100, ver) SA_RotD100 / SA_RotD50
   3. Contact Jack Baker re. Elastic/Elastic ratio (BB)
   4. Contact Jae for theoretical test

Jonney

Jason

James

1. Added
2. Added
3. Contacted and scripts ready to convert from matlab

1) NN - GMM

• Implement an initial basic pipeline with some NN config + flexible feature selection & preprocessing

2) Claudio to get familiar with classification N-net

1. Validate A18
2. Empirical (Shahi, baker model)

1. Jason
2. James

1. ?
2. Added

Abrahamson 2018 Verification

1. Sbatch : push and enforce (low priority)
2. SimAtlas simulation+animation (3rd batch)
3. e2e test on KISTI/ TACC
4. Find method for core hours usage (TACC, KISTI)
5. Dashboard missing data
6. Vs30 investigation
7. CUDA improvements.
   
   

Sung

Jonney

Viktor

2. Batch 3 done. automated workfllow not fully tested. ( should test with batch 4)

3. Almost. Fixed floating point calc issue in EMOD3D affecting TACC/KISTI. Doc for Robin to be completed shortly

4. Done

5. Done (Code changed to insert data for a specific date. Working out cumulative usage on-the-fly)

6. Preliminary investigation of vs30 tools. Extracted geo categories for all nz real stations.

7. detailed on wiki page

HF CUDA Conversion

SimAtlas simulation+animation

Tutorial: Stampede2

vs30 modifications