General Usage

The core functionalities of the application are:

• ElasticFDSG.dim2.runsim() for 2D simulations
• ElasticFDSG.dim3.runsim() for 3D simulations.

To run a simulation, the following inputs are required:

(1) a velocity model stored in a julia (.jld2) or numpy array (.npy/npz)

(2) a configuration.yaml file specifying simulation parameters.

Once the inputs are ready, a simulation can be initiated like:

using ElasticFDSG

# paths
path_to_configfile = "path/to/my/config_file.yaml"
path_to_velmodfile = "path/to/my/velocity_model.jld2"

# run a 2D simulation
ElasticFDSG.dim2.runsim(path_to_configfile, path_to_velmodfile)

# run a 3D simulation
ElasticFDSG.dim3.runsim(path_to_configfile, path_to_velmodfile)

After the calculations are completed, the results will be saved as a .h5 file at the specified location. These results can then be processed using any tool of choice that supports HDF5. Below is an example of how to access the content of the results file with Julia:

using HDF5

# extract content from result file 
function extract_hdf5_content(file_path::String)
    file = h5open(file_path, "r")
    result = Dict()
    
    function process_group(group, prefix="")
        for name in keys(group)
            path = joinpath(prefix, name)
            obj = group[name]
            if obj isa HDF5.Group
                process_group(obj, path)
            elseif obj isa HDF5.Dataset
                result[path] = read(obj)
            else
                ;
            end
        end
    end
    
    process_group(file)
    close(file)
    return result
end;

file_path = joinpath(@__DIR__, "results.h5"); # add the actual path here
content = extract_hdf5_content(file_path);

println("Content keys")
for (key, value) in content
    println("$key")
end

In the next sections, the process for creating required velocity models and configuration files will be demonstrated.