Modelling Rivers

River

Rivers are natural water courses. They have vital contributions to the social and economical welfare of societies living along their course. They provide continuous, usually dependable, flow and supply of water. Accordingly, centralised civilisation began emerging independently along number of river valleys; e.g. the Nile, Tigris and Euphrates, the Indus and the Yellow Rivers. Since then, mankind has been in continuous struggle to understand and live in harmony with this gift/force of nature. Modelling rivers is the modern way to understand their behaviour and is an essential tool to tackle several river problems.

Delft3D is the state-of-the-art modelling framework for hydrodynamics, water quality, ecology, waves and morphology.
 
Here, in this space, we share knowledge & experiences, and discuss issues related to modelling of rivers.
 
 

** PLEASE TAG YOUR POST! **

 

 

 

Sub groups
D-Flow Flexible Mesh
DELWAQ
Cohesive sediments & muddy systems


 

Message Boards

Flooding

AP
Ari Posner, modified 8 Years ago.

Flooding

Padawan Posts: 31 Join Date: 5/2/11 Recent Posts
I'm trying to simulate a flooding event in a river with a relatively small and somewhat shallow main channel and a large vegetated floodplain. When I use a flow rate that should only fill the main channel, the flow is going in to the floodplain. I am using open boundary specifications with no discharge conditions specified. I also began with uniform depth initial condition that fills the entire domain.

Anyone have advice on how to represent the wetting/drying process?
thanks
Mohamed Yossef, modified 8 Years ago.

RE: Flooding

Youngling Posts: 23 Join Date: 2/23/11 Recent Posts
Hi Ari
Firstly, I would like to know what kind of boundary conditions did you use, you indicate " I am using open boundary specifications with no discharge conditions specified".
Secondly, could you please confirm that you are using a constant depth, could it have been constant water level.
I think your problem is related to start-up procedures rather than flooding and drying.
When starting-up from a situation with constant water level, all the model will be filled with water up to this level including the floodplains. If there is a possibility for water to drain from the floodplain, your simulation should reach a state with dry floodplains.
In general, when starting a simulation from a cold start, attention should be paid to the way the model is filled or emptied from water (using something time). Sometimes, a small time step is used. After the first simulation has run for a long enough period to reach a steady state case, you should be able to create a restart file and use it for subsequent simulations.

I hope this is helpful, and would like to hear how did your simulations go.

Mohamed Yossef
AP
Ari Posner, modified 8 Years ago.

RE: Flooding

Padawan Posts: 31 Join Date: 5/2/11 Recent Posts
Mohamed,
Thanks for your interest.

My boundary conditions are:
Upstream: A total discharge time series
Downstream: A Q-H water level relation with a relaxation factor of 0.5

My initial conditions;
Uniform Water Level

I am confused about why the bed level figure in QUICKPLOT has the values as negative. Making the values negative is the only explanation I can come up with as to why the water depth values appear to be greater on the floodplains than the main channel.

Have you encountered this?

Thanks,
ari
Richard Measures, modified 8 Years ago.

RE: Flooding

Jedi Knight Posts: 178 Join Date: 3/23/11 Recent Posts
Ari,

I think you nmay have made a mistake specifying your bathymetry. The bed levels have to be specified as depths (below datum) rather than elevations (above datum) - This is counterintuitive if you are familiar with most other river modelling packages but I guess Delft3D is set up this way because it makes sense if you are working on coastal models.

I originally made this mistake with a river model and had results similar to what you are getting now. It is easily fixed if you open your *.dep bathymetry file in quickin and multiply all the depths by -1. In my cases this resulted in 'depths' of approximately -100m i.e. -100m below datum = 100m above datum = correct.

Cheers,

Richard
Bert Jagers, modified 8 Years ago.

RE: Flooding

Jedi Knight Posts: 201 Join Date: 12/22/10 Recent Posts
Hi Ari, Richard,

Indeed Richard is right. Delft3D requires the bed levels to be specified as "bathymetry" or "depth"; what is actually meant the location of the bed measured positive down relative to some horizontal reference plane. For coastal applications with a reference level at mean sea level that approximately corresponds to the bathymetry or water depth; a historical choice. Getting the sign wrong is a very common mistake for river researchers starting to use Delft3D; however, also the more experienced user needs to think carefully to interpret the bed levels correctly.

Since QuickPlot was developed by a river engineer, it tries to break this confusing historical convention and shows consistently "bed levels" that is the elevation of the bed measured positive up relative to the horizontal reference plane. Hence the sign difference compared to the model input.

Best regards,

Bert
AP
Ari Posner, modified 8 Years ago.

RE: Flooding

Padawan Posts: 31 Join Date: 5/2/11 Recent Posts
Thank you for your responses. I have resolved the bathymetry issue with your help.

I am currently trying to run a steady state simulation to test the model and create an initial conditions file for the unsteady simulation. I have a couple of questions regarding the model.

1) There are many warnings that the velocity change is too high (>5 m/s) in cells where u1 and/or v1 is 0.0.

2) I am not clear on what the difference is in using a map or restart file from the manual.

3) The tecplot export option in Quickplot does not work properly.

Have you encountered these issues?

Thank you in advance,
ari
Mohamed Yossef, modified 8 Years ago.

RE: Flooding (Answer)

Youngling Posts: 23 Join Date: 2/23/11 Recent Posts
1) There are many warnings that the velocity change is too high (>5 m/s) in cells where u1 and/or v1 is 0.0.
Normally, there are many warning at the start of the simulation. They should disappear after reaching a steady state solution. Sometime, couple of points still have this warning, just check does it affect the results.

2) I am not clear on what the difference is in using a map or restart file from the manual.
The map-file, is simply the output file of an earlier simulation and you can use it to continue the simulation from any time that is written in the map-file. The restart file is a file that includes initial (hydrodynamic) conditions for a proper start of the simulation (hot start).

3) The tecplot export option in Quickplot does not work properly.
I have never used this option.

Mohamed Yossef