Grid and wave set up - Forum - XBeach
bug message reminder
Grid and wave set up
I am sorry to beat a dead horse, but I have read and reread threads and manuals and still thoroughly confused by the grid set up wrt alfa, thetamax/min, etc. Basically, I have an east facing beach and west wave. So if thetanaut=1, alfa is ignored and so mainang = 90, but with thetanaut = 0 then alfa must be 180. So then from what reference should the thetamax and min be created? In the default case, as drawn in the grid tutorial, the thetas are drawn from the reference of the x in the east, So would their inverse be drawn from the west in the rotated case for cartesian or does it get drawn clcokwise? And for nautical, is it switched to a south origen or would both thetamax and min run counterclockwise?
I had similar problems when I started to work with XBeach. By now, I'm far from being an expert, but I might be able to help you. Personally, I prefere the use of the nautical convention (thetanaut = 1) so N = 0°, W = 270°. Applying this to your model, it follows that the wave angle corresponds to 270°. The definition of the wave grid now depends on the variety of the input wave angles and on the degree of wave interactions (e.g. diffraction) you want to allow/simulate. The width of the wave grid (thetamax - thetamin) should be sufficnently large. If, for instane, your input wave angles varied between 260 and 280°, I would start with thetamin = 210° and thetamax = 330°. The parameter dtheta now defines how many wave bins you want to have in your model:
number of wave bins = (thetamax - thetamin)/dtheta. So, if dtheta = 40° --> 3 wavebins; if dtheta = 120° --> 1 wavebin and, hence, no wave interactions.
You can check whether your wave grid is sufficiently large by looking into the xblog file after the simulation. If at a specific time step the amount of wave energy outside of the computational domain is large (look for this warning or check the amount wave energy outside the comp. domain, respectively), you should adjust your wave grid.
The parameter alfa only becomes relevant when you use local coordinates. Then, alfa is used for the correct projection of your results. Thus, the definiton of your wave grid is independent of alfa (If I understand it correctly, so no guarantee!). I always work with world coorditanes, so I don't have any experience with the parameter alfa...
Hope this helps!