Estuary with tidal intrusion

Question

Hi, I'm trying to simulate an estuary (the mouth of a river in the sea) but I can't find any information about the steps to follow to model the river that flows into the sea. I have already simulated the tidal hydrodynamic model in the estuary domain using MOHID Water. Then you should do a hydrodynamic model for the whole (tide + river discharge). Is there a guide with steps to follow for this type of simulations? Where can I find it? The explanations I have found are applied to MOHID Land, land drainage, urban flooding.
Thank you.

Luis Fernandes · Answer

Hi Gabriel, 
 Depending on the objectives of the simulation and if you know what the river flow is, you can do this type of simulation using the coastal model/engine (MOHID Water) in OpenFlows FLOOD. In this case, you solve the region of fresh water influence in 3D computing the river mixing with coastal waters. 
 Also depending on the size of river and area of interest you can potentially do everything using only one computational grid. In that case, you need to also add the river bathymetry to the grid and: 
 1) you have to expand your computational grid inland to accomodate the estuarine area of the river (you should simulate the river upstream to where the tide has influence) 
 2) increase the grid resolution so that the river is well represented. If this results in a grid too big you may want to refine it using a submodel. If that is the case, I can provide some additional information, please let me know. 
 3) Add the river flow as a Discharge. Add the discharge module into your simulation and specify the location and flow options. 
 Is the focus the coastal/riverine mixing area? Is this only the river? Are you looking to compute river conveyance dependency on the sea level boundary condition? There are many possibiities, so depending on the speficic details I may be able to provide more suggestions. 
 Kind regards, Luis

Luis Fernandes · Answer

Hello Gabriel, 
 Please check the article about Nesting sub-models in coastal applications in OpenFlows FLOOD . It contains a walkthrough guide with all the steps necessary to refine your solution using a nested model approach. 
 Please let me know if you have any additional questions. 
 Kind regards, 
 Luis Fernandes

Luis Fernandes · Answer

Hello Gabriel, 
 I took a look at your model setup and: 
 You've setup the "father" model hydrodynamic module to output time series results and set a path to the time series location file that does not exist 
 
 I have fixed that and ran the model. Then I got an error message related with the existence of isolated grid cells in your bathymetry files (both for the father and the son model). This results in the model generating a corrected bathymetry file. So far so good, but for some reason it seems that FLOOD is also activating an option that if this happens the simulation does not stop. To fix this you should open both the father and son Model.dat input files in a text editor and change: 
 STOP_ON_BATHYMETRY_CHANGE : 0 
 to 
 STOP_ON_BATHYMETRY_CHANGE : 1 
 Then run the simulation again and after it starts the model will stop and warn you that it has changed the bathymetry. You should then change the bathymetry file path to the newly generated file. You'll have to do it for the father and son domains, 
 After I've fixed it and ran the simulation, another error is produced: 
 ReadAssimilationFilesName - ModuleAssimilation - ERR01 
 
 This is caused by the fact that in the "son" domain Hydrodynamic module you've set to True the option Data Assimilation and Flow Relaxation. 
 When you do that you'll need to activate the Assimilation module in your simulation options and then configure it accordingly. Given your simulation consists of two 2D domains forced only by tide, you'll possibly don't even need to activate the Data Assimilation and Flow Relaxation. In case you want to do it you may want to look at this article from the MOHID Water wiki (see example 2): 
 Please let me know if this solves your problem or if you need any additional information. 
 Kind regards, Luis

Luis Fernandes · Answer

Hi Gabriel, 
 Without knowing the details and objectives of your simulation, I can't make many recommendations, but looking now in a little more detail to your grids/bathymetries I see the following: 
 1) the father grid has many areas that possibly didn't have enough bathymetry data and you've possibly chosen not to extrapolate during the bathymetry interpolation process. This produces many areas where the bathymetry is set to -55 (which is a default in OpenFlows FLOOD for undefined bathymetry value). If you play with the bathymetry color scale and plot all values below zero with a violet color you should see those areas easily, see example below (but there are at least 7-8 areas like this). 
 
 2) The son grid does not have sufficient resolution to model the upstream part of the river (see image below). The resolution should allow for adjacent grid cells to share a face. If not flow will not pass from one to the other. Also, the son bathymetry has the same problems in some areas as described in point 1) 
 
 Having more details on what you're trying to simulate I could make other suggestions .Maybe use a curvilinear grid for the river? 
 Hope this helps. Kind regards; Luis

Luis Fernandes · Answer

Hello Gabriel, 
 As stated in the error message it looks like the polygon line is not a simple polyline. Check for possible line self-intersections and try to simplify the line where possible. Note that you do not need to give a very detailed outline of the coastline. You only need to delineate the major indents (e.g. tributary rivers, small bays, etc.). 
 You may find useful information in this article about generating curvilinear grids in OpenFlows FLOOD . 
 You can always send us the polygon file so we help point out what's causing the problem and how to solve it. 
 Kind regards, 
 Luis

Luis Fernandes · Answer

Because you have some many indents in your polygon line, but you've only identified 4 "turn right" points it gets complicated for the algorithm to generate a fine resolution grid. Try simplifying the polygon line by creating a new one enveloping the river area but with less points and no sharp small turns in direction (e.g. the line looks to follow a gridded format, with sharp 90 degree turns). Please let me know if following those guidelines solves your problem. 
 Kind regards, Luis