Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

Task

Expected time reqd.

People

Notes

1. Source          (HOBY)

GeoNet Mw email to SRF

(a)  from Mw tensor get Rup parameters

(b)  From Rup parameters use createSRF.py to make rupture SRF

(c)  Execution of createSRF.py

 

1minute

5mins

 

10sec

6min total

 

Hoby

a) Manually extract location, dip,strike, rake from Mw tensor (Ristau) into setSrfParams.py

b) Change TYPE=2 (point source to finite fault)

c) Run "createSRF.py"

    

Plotting of SRF

(a)  SRF plane(s) standard plotting

(b)  SRF on map to verify location (to come after domain set in step 2)

 

3mins

5mins

2mins

10min total

 

Hoby

 

(a) Hoby(b) Viktorchange manually the SRF and output name in plotSRF.py (need to be improved)

Execute ‘plotSRF.py’ (select title for top of script)

(a) Hoby : Just needed to change SRF name 

(b) Viktor: make 2 plots on 1 image, 1 zoomed in with slip visible, 2 whole of NZ showing outline/hypocentre

Finished automating. only SRF needed, no reliance on params.py etc...b) plot_srf.py file.srf

2. Velocity Model Domain   (ETHAN/ROBIN)

From rupture location to VM dev and domain

(a)  From location and Mw determine VM origin, 3D extent, and simulation duration

(b)  Plot VM domain extent

(c)  Generate VM with H=0.1km for this domain, then extract slices

1mins

20 mins (on multiple cores)

15 mins for slices

4 mins for H=0.2km VM creation (30 mins for H=0.1km)

20min total

Ethan, Robin,…

  • Tidy up params_vel.py (some params are redundant) and include corners for plotting before mesh generation
  • Run gen_corrds during VM and place all model params alongside VM files
  • Integrate the gen_cords into the first step to enable the domain to be plotted on a map and checked / changed if necessary
  • Split the VM generation from VM interrogation / slice generation to allow the VM to be copied while slices are being generated
  • Will need a wall clock time estimation done before VM calculation
  • Merge  EXTENT_Z_SPACING, EXTENT_LATLON_SPACING and HH
  • Change names of to be consistent  ORIGIN_LAT, ORIGIN_LON,ORIGIN_ROT retain these names , (MODEL_LAT,_LON,_ROT )
  • Write xy2ll into the gen_params.py script to write the domain corners into params_vel.py
  • Plot the vm domain box within gen_params.py
  • Include gen_cords.py within genDomain to keep coords and .p .s and .d binary files in together

 

  • There was an error in the auto-VM generation on hypocentre which needed to be fixed, and then the code had to be obtained from github again.

From non-uniform station grid, get statgrid points

(a)  Locations of statgrid points

(b)  Locations of GeoNet strong motion stations

(c)  From NZ Vs30 model get Vs30 values at all points

 

(ignore for now)

1-2 hours!!

10mins

~2hours

 

 

Ahsan, Chris

 

 

[need .ll file for all of NZ (Ahsan)] (Done)

run statcords (to get station locations corresponding to actual VM domain]

(a) Plot VM domain extent, stations (statgrid and GeoNet),

(b) color based on Vs30 values

2mins

(a) Viktor

(b) Not done

Only done as part of plotting PGA, PGV, PSA, PSA RATIOS

Where to get vs30 values?

3. Observed strong motions  (SUNG)

Obtain observations from GeoNet

(a)  Get recorded motion from GeoNet ftp

(b)  Process and reformat using defaults

(c)  Produce plots of acc/vel time series, response spectra

 45 mins total for (a)

 and (b)

60 mins for generating Vs30 values. (one-off task)

5 mins for (c)

Ahsan

  

Majority of time spent trouble shooting. See comment below

GeoNet people make errors which can cause errors and breakdown our code. The data location has files that should have been removed. Take a look at ftp://ftp.geonet.org.nz/strong/processed/Proc/2013/08_Aug/2013-08-16_023105/Vol1/data . Files ending with _ should not be there and nor should V2A and V3A data be there. You will also see three stations named like Q13924J03.V1A that don't follow the naming convention used for all the other data files. This broke down my code. For the next drill, I will change the code to catch these as exceptions and continue smoothly.

Plot IM values at GeoNet stations on map (for various IMs)

3 mins

Ahsan

Calculations of PGVs, PGAs and pSA for set of chosen periods took about 3 mins, these were then passed on to Viktor who plotted them.

Plot observed vel time series on map

5mins

Viktor

skipped, need to finish script that does this

Plot IM values as a function of Rrup and compare with empirical ground motion model (various IMs)

10 mins

Ahsan

Code ran as expected.

4. Perform GM Simulation   (SUNG)

Submit GM Simulation

(a)  prepare input params.py, pointing to SRF, VM, and stations of interest etc. (gen_cords.sh and statlist2gp.py need to be run)   File transfer and install

(b)  LF sim (queue time + run time)

(c)  Once LF sim complete, run HF and BB sim

 

(a)5mins

(b)5 hours (queue+run)

(c)13 min

 

Sung

 

gen_ts was broken (fixed)

Will be nice to have auto file transfer (in 3 separate streams esp. for VM model)

Will need to separate HF from BB and run it alongside LF.

 

5. Post-processing   (VIKTOR)

LF Time slice animation

(a) tar, download data from Fitzroy to local

(b)  From TS produce PGV and MMI plot

(c)  Produce TS animation

 

10mins

25mins

25min

 

 

Viktor

 

needed to fix up scripts to work with new parameter style

mainly needed to fix script to use new parameter style, should run automatically after step A
run time only ~3 minutes 

need to automate a little more, run time ~5 mins

Script preparation and test

At SMS locations compare sim vs. obs

(a) compare waveforms

(b) Compare IMs as a function of Rrup (multiple IMs; sim, obs, empirial)

(c) Compare sim, obs, empirical response spectra at individual locations

(d) Plot overall bias as a function of vibration period (both sim and empirical)

(e) Preparing input for (f)

(f) Plot bias(residual) at specific locations on a map

10mn

(a)–(c) 30 mins(d) 30 mins

 

 

 

 

 

 

 

 

(e) 4 mins

(f) 5 mins

Hoby

 

 

 

 

 

 

 

(d) 20mn 

 

 

(e) 10 mins 

(f) 5 mins

Hoby

Viktor


  • Testing the Script using only the observation and remove irrelevant stations
  • Make sure to use fd_*.ll station lists instead of geonet*.llll station lists instead of geonet*.ll
  • Need to add Ground motion durationThere were some errors during the bias plotting, so I need to remove that part of the code to plot the IMs

 

 

 

 

default parameters used, runs in 2 mins but needed to fix a few things up

...

 TasksTime estimation
HobyFix bias plotting for both empirical and physics-based models

1) Include ground motion duration in the PP plotting

2) Add the empirical ground motion duration

2) get plotSRF improved

2days

2days

2-5 days

2days

Ethan
1) Streamline the VM generation with gen_coords.csh
2) Write scrip to stitch together multiple PDFs in PDFtk
3) VM domain baseplot to be passed to plot SRF by others
2days (with other minor tasks)
Ahsan (1) GeoNet people make errors which can cause errors and breakdown our code. The data location has files that should have been removed. Take a look at ftp://ftp.geonet.org.nz/strong/processed/Proc/2013/08_Aug/2013-08-16_023105/Vol1/data . Files ending with _ should not be there and nor should V2A and V3A data be there. You will also see three stations named like Q13924J03.V1A that don't follow the naming convention used for all the other data files. This broke down my code. For the next drill, I will change the code to catch these as exceptions and continue smoothly.

(2) Deal with new stations better. Currently I maintain a list of about 1350 stations and new stations coordinates are read from there. But in this drill one new station was not in that extensive list. A better solution would be to read the data file and extract the coordinates rather than relying on a list.

(3) I shall add Vs30 = 500 for any new stations and give the list of new stations to Kevin after our drills. He can then provide a better Vs30 estimate. This way we simply the problem of knowing Vs30 for new stations. What do you propose?

These are not big issues in general but they are if we want unbroken, smooth work flow during real time simulation. I shall add these to my task list.

(1) 2 hours

 (2) 5 hours

(3) 2 hours

SungThe following 3 items are easy to automated, and where we can easily make mistakes.
  1. Create shared parameter files that can be imported into other steps in the workflow (Saves need for configuring params.py and get_cords.sh)
  2. Automated file transfer to Fitzroy
  3. Automated installation of a job

  1. 2 days
  2. 2 days
  3. 2 days
Viktor
  1. Finish script to plot observed timeseries on map
  2. Run plots automatically when data has been retrieved
  3. Determine format and implement for plotting stations on map with PGV, PGA, PSA data etc...
    and figure out easiest way to plot PGV stations on overlay

2.5 days
1 - 2 days
2 days 

Daniel
  1. Non-uniform grid should probably run after the VM
  2. Include non-uniform grid generation to the automatized workflow
  3. Parameters for the grid should be read from the VM file
  4. Make sure the format of the output is usable
 

...