Testing GeoClaw

[Luis Montoya]

Hi,

 

My name is Luis and I am currently a PhD student at USC. This is my first time using GeoClaw and I am running into some problems. I will be using GeoClaw for tsunami modeling. I have already installed the Clawpack packages for version 5.3.0. I have tested my installation under the classic/tests and amrclaw/tests folders and no errors were encountered. But, when I try  to test GeoClaw (geoclaw/tests) I am getting errors in all the subfolders (i.e storm_surge, etc). I am typing the command make .plots or make all and I am getting the following errors:

touch /home/../auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/src/2d/shallow/surge/storm_module.mod; gfortran -c -cpp /home/../auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/src/2d/shallow/surge/storm_module.f90 (….continues few more lines)

 

type(constant_storm_type) :: constant_storm

                                               1

Error: Object 'constant_storm' at (1) must have the SAVE attribute for default initialization of a component

/home/../auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/src/2d/shallow/surge/storm_module.f90:49.45:

 

    type(holland_storm_type) :: holland_storm

                                             1

Error: Object 'holland_storm' at (1) must have the SAVE attribute for default initialization of a component

make: *** [/home/../auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/src/2d/shallow/surge/storm_module.mod] Error 1

Thanks for the help.

 

Luis

[Kyle Mandli]

Hi Luis,

That is a bug (my fault actually) which pops up on certain versions of gfortran and the intel fortran compilers. Luckily the fix is easy, add the `save` attribute to the lines below and everything should work, i.e.

> type(constant_storm_type), save :: constant_storm

and

> type(holland_storm_type), save :: holland_storm


You also might want to add it to the other storm type there just in case

> type(holland_storm_type), save :: stommel_storm


I also (finally) created a PR with the corrections (https://github.com/clawpack/geoclaw/pull/160) if you want to take a look at what changed. You could grab the new changes once this is merged if you are comfortable using git and the development version of the code.

Kyle

> --
> You received this message because you are subscribed to the Google Groups "claw-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to claw-users+...@googlegroups.com.
> To post to this group, send email to claw-...@googlegroups.com.
> Visit this group at http://groups.google.com/group/claw-users.
> For more options, visit https://groups.google.com/d/optout.

[Luis Montoya]

 

Thanks for the help Kyle.

 I am now trying to run a simple tsunami case but no output was created (except the  .data files under the _output directory). I am using a 3 min grid (Level1_3min.asc) topotype=3, and my initial condition is a surface elevation file (Disp_Tohoku.asc, format topotype=1) iqinit = 4. I attached the files to this message.  I read the tsunami example on the website and tried to follow the same logic but when I run it no output is produced. Here is what I am getting:

 

 

Makefile:187: warning: overriding commands for target `all'

/home/../auto/wave/lmontoy/geoclaw/clawpack-5.3.0/clawutil/src/Makefile.common:259: warning: ignoring old commands for target `all'

make output -f  Makefile /home/../auto/wave/lmontoy/geoclaw/clawpack-5.3.0/clawutil/src/Makefile.common

make[1]: Entering directory `/auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/examples/TOHOKU/chile2010'

Makefile:187: warning: overriding commands for target `all'

/home/../auto/wave/lmontoy/geoclaw/clawpack-5.3.0/clawutil/src/Makefile.common:259: warning: ignoring old commands for target `all'

rm -f .output

python /home/../auto/wave/lmontoy/geoclaw/clawpack-5.3.0/clawutil/src/python/clawutil/runclaw.py xgeoclaw                  _output                \

        True False . False False None

==> runclaw: Will take data from  /auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/examples/TOHOKU/chile2010

==> runclaw: Will write output to  /auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/examples/TOHOKU/chile2010/_output

==> runclaw: Removing all old fort files in  /auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/examples/TOHOKU/chile2010/_output

 

==> Running with command:

    /auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/examples/TOHOKU/chile2010/xgeoclaw

Reading data file: claw.data

         first 5 lines are comments and will be skipped

Reading data file: amr.data

         first 5 lines are comments and will be skipped

Reading data file: regions.data

         first 5 lines are comments and will be skipped

Reading data file: gauges.data

         first 5 lines are comments and will be skipped

Reading data file: fgmax.data

         first 5 lines are comments and will be skipped

 

 Running amrclaw ...

 

Reading data file: geoclaw.data

         first 5 lines are comments and will be skipped

Reading data file: refinement.data

         first 5 lines are comments and will be skipped

Reading data file: dtopo.data

         first 5 lines are comments and will be skipped

Reading data file: topo.data

         first 5 lines are comments and will be skipped

 

 Reading topography file  /auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/examples/TOHOKU/chile2010/Level1_3min.asc                                          

 **** topo arrays do not cover domain

 **** area of overlap =    0.0000000000000000

 **** area of domain  =   -223.50249999999988

 

==> runclaw: Finished executing

 

==> runclaw: Done executing /auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/examples/TOHOKU/chile2010/xgeoclaw via clawutil.runclaw.py

==> runclaw: Output is in  /auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/examples/TOHOKU/chile2010/_output

make[1]: Nothing to be done for `/home/../auto/wave/lmontoy/geoclaw/clawpack-5.3.0/clawutil/src/Makefile.common'.

make[1]: Leaving directory `/auto/wave/lmontoy/geoclaw/clawpack-5.3.0/geoclaw/examples/TOHOKU/chile2010'

[Luis Montoya]

I forgot to attach the setrun.py file, find it attached. Thanks

Luis

...

[kyle....@gmail.com]

From the output it looks like GeoClaw is not reading in your topography/bathymetry correctly as it thinks the area of your domain being covered by your grid is negative.  This is usually a sign that the format of your bathymetry file has an error, in particular the order of points.  Checkout the info at 

http://www.clawpack.org/topo.html

and see if that helps.

Kyle

<Disp_Tohoku.asc><Level1_3min.asc>

[Luis Montoya]

Hi Kyle,

I am trying to use the fixed grid monitoring feature but I am a bit confused. I am trying to run a tsunami simulation in Japan using 5 nested grids. I would like to get the max vel, mom and surface elevation values for my finest grid. How do I specify this in these lines?


    # == setfixedgrids.data values ==
    fixed_grids = rundata.fixed_grid_data
    # for fixed grids append lines of the form
    # [t1,t2,noutput,x1,x2,y1,y2,xpoints,ypoints,\
    #  ioutarrivaltimes,ioutsurfacemax]

Also how do I use the .py routine to generate the fgmax_grid.txt? Or is it created automatically once I compile the files?

Thanks,

Luis Montoya

[Randall J LeVeque]

The lines from setrun.py you quote are for the version of fixed grid output that has been deprecated, see

  http://www.clawpack.org/fgout.html

Instead we recommend using what's described at

  http://www.clawpack.org/fgmax.html

There are a couple examples of usage in the apps repository, see

  http://www.clawpack.org/gallery/gallery_geoclaw.html#fgmax-examples

  http://www.clawpack.org/apps.html

See setrun.py and make_fgmax.py in these directories, e.g. $CLAW/apps/tsunami/chile2010_fgmax

[Kyle Mandli]

Hi Luis,

I unfortunately do not have a lot of experience with the fixed grid capabilities in GeoClaw and the documentation for 5.3.x is out of date at this point (http://www.clawpack.org/fgmax.html?highlight=fixed%20grid). Randy LeVeque will probably be the person to answer your questions beyond my brief knowledge but I will suggest looking at the examples in https://github.com/clawpack/apps/tree/master/tsunami as they both use the updated fixed grid support.

Kyle

[Luis Montoya]

Thank you Prof. LeVeque

[Luis Montoya]

Thank you Prof. Mandli.

I am now having trouble plotting my results using python. Everything goes smooth until I use the plot command. This is my first time using python and I have been doing a lot of reading to learn how to use it. I am getting this error:
>>> topo.plot()
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/storage/home/lhmontoy/GeoClaw/clawpack-5.3.0/clawpack/geoclaw/topotools.py", line 861, in plot
    fig = plt.figure(**fig_kwargs)
  File "/storage/home/lhmontoy/anaconda/lib/python2.7/site-packages/matplotlib/pyplot.py", line 435, in figure
    **kwargs)
  File "/storage/home/lhmontoy/anaconda/lib/python2.7/site-packages/matplotlib/backends/backend_qt4agg.py", line 47, in new_figure_manager
    return new_figure_manager_given_figure(num, thisFig)
  File "/storage/home/lhmontoy/anaconda/lib/python2.7/site-packages/matplotlib/backends/backend_qt4agg.py", line 54, in new_figure_manager_given_figure
    canvas = FigureCanvasQTAgg(figure)
  File "/storage/home/lhmontoy/anaconda/lib/python2.7/site-packages/matplotlib/backends/backend_qt4agg.py", line 72, in __init__
    FigureCanvasQT.__init__(self, figure)
  File "/storage/home/lhmontoy/anaconda/lib/python2.7/site-packages/matplotlib/backends/backend_qt4.py", line 68, in __init__
    _create_qApp()
  File "/storage/home/lhmontoy/anaconda/lib/python2.7/site-packages/matplotlib/backends/backend_qt5.py", line 138, in _create_qApp
    raise RuntimeError('Invalid DISPLAY variable')
RuntimeError: Invalid DISPLAY variable

Luis

[Kyle Mandli]

It looks like QT (one of the backends for matplotlib) is not finding the windowing session. What kind of system are you on and is it headless (no attached screen)? You could also try switching to the TkAgg backend which is pretty reliable although this may not solve your windowing issue (I would suggest modifying your matplotlibrc file to do this although you can do this via a python session, see http://matplotlib.org/users/customizing.html).

Kyle

[Luis Montoya]

Kyle,

I am connecting through the terminal to a Linux system (which has no attached screen).

Luis

[Kyle Mandli]

You can change the matplotlib backend to Agg in the way I mentioned in the previous response. This will not allow interactive sessions but will allow you to view the plots. Alternatively you can setup your ssh connection to use the display on the local machine to do draw the windows. You will have to google around a bit for how to get this to work as it depends on the platform unfortunately.

Kyle

[Luis Montoya]

Thanks a bunch Kyle! It is working perfectly

[Luis Montoya]

Hi Kyle,

I am trying to run the Tohoku tsunami event using 5 nested grids. My simulation just finished but in the fort.FG1.valuemax file I am only getting -0.99E+99 for every grid point. I specified in the make_fgmax.py file that I want to monitor the max on my level 5 grid (fg.min_level_check = 5    # which levels to monitor max on) but I noticed that in the output.log file that calculations are being made only in Levels 1-3. Do I have to make two separate runs for this many grids? Thanks

Luis

[Luis Montoya]

Hi Kyle,

I am trying to plot the results from a GeoClaw tsunami run in Matlab. I know that there are many scripts available to do this but I just want to create a matrix with my z (water level) values from the fort.q files. This way I can analyze certain areas I am interested in. Is there a easy way to do this?  Thank you

Luis

[Kyle Mandli]

Hi Luis,

The fort.q files from GeoClaw do contain the sea-level heights although if you really want this in a particular location on a grid you may want to use a fixed-grid so you can specify specifically where the grid is you are interested in. This would simplify the output and creation of your matrix you want.

Kyle

[Luis Montoya]

Thanks Kyle!

Is there a way to turn off refinement from the setrun.py file? Or do I have to do it in the code (Fortran files)?

Luis

[Kyle Mandli]

Hi Luis,

Yes, set `amrdata.amr_levels_max = 1` which will instruct the code to only use one level effectively turning off refinement.

Kyle

[Luis Montoya]

Hi,

I am trying to run the recent 2015 Chile event and I am getting this error:

*** Error, grid cell does not overlap any topo grid
  xim =     0.29188671832623E+03  xip =     0.29195335916312E+03
  yjm =    -0.73960000000000E+02  yjp =    -0.73912775269191E+02

Does this mean that one of my level grids does not overlap with the others?

Thanks,

Luis

[Kyle Mandli]

Hi Luis,

It means the topography you provided does not cover the entire domain you requested.

Kyle

[Luis Montoya]

Hi Kyle,

Thanks!

Luis