Hi,

I am currently modelling the 3D propagation of a GHz ultrasound wave in the context of picosecond ultrasonics. I have a point source and sensors along a single dimension of propagation. I have a grid of 1024x512x512 (grid spacing 0.048828125um in all directions) that I am running with the C++ open MPI code on an HPC. I had determined the timestep and t_array using the maketime function.

The attenuation coefficient is set to 0.0022 and the power is 1.01

I was noticing that there is no significant change in the frequency spectrum of this wave as it propagates through space. A spect of the wave returns similar (albeit not identical) fourier spectrums at a distance of 5 um away from the source and also at a distance of 46um away form the source. Now This isn't necessarily a problem, but I just wanted to make sure that these results are trustworthy. I was expecting that the fourier spectrum at this distance would look drastically different, with the higher frequencies having largely diminished. Is the power law absorption model valid for propagation at such frequencies? If not, is there a fix to this problem or any other way to model GHz frequency attenuation?

Thanks,