Alan Bowers Design Engineer / Hydraulic Modeler

Have you tried to combine this with suggestion from the other post to fix negative water levels?

That's great!  You are welcome.

Sam,

From the 2014 M21HD manual, section 6.12.3 on Inland Flooding:

[quote]In certain situations MIKE 21 may yield negative water depths. This can
occur for instance in areas with steep gradients in the bathymetry (or
topography) or by a relatively large time step in the model. To minimize
the potential for mass falsification you should generally use smaller flood
and dry levels for inland flooding. Appropriate values in inland flooding
could be Drying depth in the range 0.001 - 0.02 m and Flooding depth in
the range 0.002 - 0.04 m, where Flooding depth > Drying depth.
NOTE: The drying and flooding depth must be larger than 0.0001 m and
0.0002 m, respectively.
[/quote]

Hope this helps,
ab

Sam,

What flow behavior is the model exhibiting?  How does this differ from what you expected?

I would infer that you are getting pooling instead of flow leaving the domain.  I neglect to use a 'Boundary' at all for my M21 overland flow simulations.  Instead, I create an artificial boundary by manually raise the elevation at all of the boundary cells to something arbitrarily large to contain any flow (or else you will get an error) and place 'Sinks' on the cells where you want an outflow boundary, setting the 'Magnitude' to something arbitrarily large to allow a free outfall.

I'm sure there is a way to do this with the boundary approach, but like many other aspects of this software, I'm not sure how.  Maybe you will find this workaround useful.

ab

Ali,

I am not familiar with non-uniform grids in MIKE, but I have experience with non-uniform 3-D meshes in other software.  What exactly are you attempting to model?  Can you include a screencap of the grid?  It is difficult, otherwise, to diagnose numerical instability.

My initial guess would be that your grid and/or time discretization are too large for the flow velocity you are attempting to model (see Courant Number, CFL criteria).  What values are you using?

Instability could also be caused by poor mesh/grid quality.  I'm not sure what freedoms MIKE gives you for non-uniform meshes, but an abrupt 'jump' in cell size can bring about numerical error, especially if a large gradient encounters such a cell (in this case, position or velocity, but the same is true with heat, pressure, shear stress, etc.)

Hope this helps.

Samuel,

The program uses an Alternating Direction Implicit solver that can produce oscillations, although I wouldn't expect them to be so pronounced with reasonable settings (see pg. 13 of the M21HD_ScientificDoc). 

What time step and grid size have you used?  I would suggest cutting each in half (maybe more if they were unreasonably large to begin with) and observing the effect it has on your results, if you haven't already.  If the oscillations disappear or are significantly reduced then I would conclude it was just numerical error of the solution scheme.

Regards

Attached are zoomed views of flow depth on time steps 15 and 16 and the associated hydrographs. 
The hydrographs show flows as early as time step 9 and all are clearly staggered and should not be appearing at the same step. 
These are applied at each point from (558,1849) - (705,1849).

Hello all,

I am condensing a rather large regional MIKE model into a smaller scale model closer to the domain of interest.  This is accomplished by using the MIKE discharge tool to extract flow (m^3/s) from the regional model and assembling every cross-sectional discharge into an input hydrograph for the small-scale model. 

However, the results from the new small-scale model do not match the inflow hydrograph that was applied.  Cells that should be wet are dry for many time steps and then suddenly all of the inflow nodes are wet (at too late a time step), as if they had all been turned on at the same time.  I have reduced flooding/drying depths and double and triple checked that the correct data is being used and applied to the correct nodes.

I'm looking for any insights into why the inflow hydrograph isn't being exhibited in the results as expected.  Any thoughts or suggestions are greatly appreciated.

Thanks!

Hello,

I have a single .dfs2 output file containing water depth, x-flux, and y-flux.  I am familiar with the MIKE ZERO tool MIKE2GRID, which exports the water depth data to .asc, but how can I similarly export the other variables, x-flux and y-flux?  Using Release 2014, SP 1.

Thanks.