Creating a Tsunami Simulation Program: Part 3

The previous part is here.

I skipped the details regarding the terrain data, so I will cover them again.
When I opened the terrain data in a text editor, it looked like the following (reposted).

The cell size was 0.0041666666666….
This is the value in latitude and longitude.
Since the GEBCO data was 15-second mesh data, I will verify that.

15 seconds = 1/4 minute
1 minute = 1/60 degree

Therefore, the 15-second mesh is 1 ÷ 60 ÷ 4 degrees.
1/60/4 = 0.00416666666…, which is correct. Just confirming.

Also, while the xcorner is 120 degrees, in the data output as text from the netCDF file using the GMT grd2xyz command, the x-coordinate of the grid position is '120.00208'. 0.00208 is 0.0041666/2, so it seems that the center coordinates of the mesh where the elevation values are given are being output.

Now, when I created the simulation program previously, I performed calculations with a time step of 1 second, but when I ran the calculation as is, it diverged. (The amplitude became infinity.)

It seems that because the mesh size became smaller, the numerical calculation diverged as it was. Therefore, after some trial and error, I reduced the time step to 0.2 seconds. In exchange, the calculation time required to perform a simulation for a fixed time after the earthquake occurred has increased. This is because the number of calculation time steps has increased.

Now, last time, I showed the simulation results of the progress so far.

The tsunami height is shown with color contours, but since it is difficult to see the shape of the fine waves, I recreated the video by adding shading. The result is shown below.

When zoomed in, the run-up onto land in Miyagi and Fukushima is also reproduced. It seems the accuracy is higher than what I created before.

The source model I am currently using for the calculation is a rough one with uniform slip, so I would like to perform a comparative verification with actual tsunami data using a more realistic source model in the near future.