Message Boards

Back

Delwaq2 NaN value for continuity-check

GS
Gholamreza Shiravani, modified 3 Months ago.

Delwaq2 NaN value for continuity-check

Youngling Posts: 21 Join Date: 6/25/16 Recent Posts

Hi,

After running Delwaq2, the continuity value becomes NaN in some part of Domain. Could you tell me the reason of that? When the Delwaq2-parameters beom NaN (None a Number)?

Michelle Jeuken, modified 2 Months ago.

RE: Delwaq2 NaN value for continuity-check

Jedi Knight Posts: 156 Join Date: 1/21/13 Recent Posts

Hi,

 

This could not be due to any process parameter, since no processes are involved. It could happen when you have used a transport scheme that is not appropriate for the case you are modelling and/or when your time step is too big. Try using an other scheme and/or a smaller time step. High vertical dispersion could also cause problems. You could apply a factor on the vertical dispersion from FLOW.

 

Michelle

GS
Gholamreza Shiravani, modified 1 Month ago.

RE: Delwaq2 NaN value for continuity-check

Youngling Posts: 21 Join Date: 6/25/16 Recent Posts

Hi Michelle,

Time-step and numerical methods are not the reasons, because I tried implicit methods (such as 15 or 16) as well as explicit methods (e.g. 11) with different small time-steps (like 5 seconds!) but the NaN problem is still printed in .mon-file. 

Using the .ada-file in a map-view, I realized that  NaN values are started from the dry and wet areas (the tidal part of the Domain(e.g. an island)) and expanded to rest parts of the domain till all cells become NaN. Has Delwaq a problem with dry/wet boundaries? I have in my open-boundary an island, where is included in my D-Flow and using bathymetry elevations behaves on its boundaries as dry/wet boundaries. But in Delwaq, these areas become NaN and cause to all cells after some time-steps become gradually NaN. Could you tell me how can I solve this problem?