IS1 not contributing to action density spectrum

[Biao-Zhao]

In w3srcemd.F90, it appears that the ice scattering source term calculated by IS1 is not contributing to the increment of the action density spectrum (SPEC).

From the two screenshots above, we can see that ATT expects VDIR from the call to W3SIS1. However, W3SIS1 does not return it.

From an energy conservation perspective, even if W3SIS1 were to return VDIR, I believe it would still be inappropriate to consider the ice scattering term as dissipation. Based on my understanding, it might be better to redistribute SPEC in different directions, following the same approach as IS2. Alternatively, we could include VSIR in VS, such as:

VS(IS) = VS(IS) + VSIR(IS)

Anyway, I might have misunderstood some details, and I appreciate any corrections or clarifications. Please feel free to share your thoughts.

[NickSzapiro-NOAA]

@cmbitz reports the same in #1305

[Biao-Zhao]

@cmbitz reports the same in #1305

Hi Nick,

Thanks for sharing the information. I hadn't noticed it before. Personally, I’m not particularly interested in wave-ice interaction, but I came across this bug while reviewing

w3srcemd.F90

As @cmbitz said, it is easy to fix it. I shared this here as a reminder for anyone working on this topic. Since scattering doesn’t dissipate wave energy, I believe the correct solution is to redistribute the scattered energy to other directions, similar to IS2. In this sense, adding the following line should solve the problem

SPEC(1+(IK-1)*NTH:NTH+(IK-1)*NTH) = SPEC(1+(IK-1)*NTH:NTH+(IK-1)*NTH) + VSIR(IS) * DTMIN

Of course, the following line must be commented out accordingly. Not only because it doesn’t work, but also for physical considerations

`!/IS1             ATT=ATT*EXP(ICE*VDIR(IS)*DTG)`

If anyone thinks it’s necessary, I would be happy to submit a pull request