Stefan Leschka

Hi Luis,

beside the very good suggestions by Igor, pointing out mesh quality which governs the quality of the solution and also the stability, here a few other challenging situations which you can try to prevent:

* flooding and drying at boundaries
* large differences in two neighbouring boundary conditions (e.g. difference in water levels close to each other)

Cheers
Stefan

There are 2 ways:

1) Use Matlab or Python for loading, modifying and writing dfsu files. Among other tools which might come in handy, the dfs libraries for free download here: https://www.mikepoweredbydhi.com/download/mike-by-dhi-tools

2) If manual modifications would be feasible, then I suggest to think of it as a 2-step procedure: (i) Run the simulation with constant evaporation/precipitation and write the items as outputs. (ii) modify the output file so that it serves as input for the next (final) simulation.
When doing (i), make sure that in the output file you have the right number of timesteps. You could do a short simulation with very short timesteps and then change the time step size of the output in the "Data utility". To find Data Utility: (a) open MIKE Zero, (b) click on "File"->"Options"->"Data Utility..." and (c) load the dfsu file. Once the file is prepared, you would need to modify the values in the different timesteps of the dfsu file.

Hope it gets you one step further.

Stefan

Hi Vadlamudir,

please note that the numerical methods in both modules are different (finite differences in MIKE21 vs. finite volumes in MIKE21FM). When switching the model from MIKE21 to MIKE21FM, you need to recalibrate it.

Hi Daniel,

could you solve your issue already?

On the source "size": It has to be a point source. To cover an area, one way would be to setup (standard or simple) source per element.

Best regards
Stefan

Hi Daniel,

which MIKE version and which boundary condition are you using?

Best regards
Stefan

Hi,

yes, recently, the rotation angle is locked. You could change the *.grd file itself so that it represents an oblique wall.

Hi Vadlamudi Premkumar,

did you try specifying a discharge at the upstream boundary and a water level at the downstream boundary? You may also try free outflow at the downstream boundary.

Cheers
Stefan

Hi,

a sink is a negative source. Put in a source and specify the amount which you want to take out with a negative sign. See also here: https://manuals.mikepoweredbydhi.help//2021/Coast_and_Sea/MIKE_21_Flow_FM_Scientific_Doc.pdf

Cheers
Stefan

Hi Vadlamudi Premkumar,

you would usually have multiple boundary conditions. Not sure what you mean. Can you specify? It would also be good to know which kind of model application you are looking at (long river, lake, or?).

Best regards
Stefan

Hi Muserica,

Assuming that you have the resistance values (Manning's n) with geographical reference, you can interpret them as "bathymetry" data and use the Bathymetry editor, which can interpolate the resistance values onto a dfs2 grid.

Best regards
Stefan