Hydraulic obstacles on drainage network

[cyrilgarneau@gmail.com]

Hi,

I'm working on a highly anthropized river and I have three questions:

1. There is a weir in the middle of my river. I can simulate it with a steep gradient in the river slope, but its presence also affects the erosion. As the weir is in concrete, no erosion should happen. Can I modify the erosion parameters for one node? This weir also generates numerical instability in my model: Under a constant flow, the cohesive sediment fluctuates from 0 to 40 mg/l with a period of 2 hours. The LIMIT_DT_COURANT and MAX_COURANT keywords helped to remove the instability in the flow. But they don't act on the cohesive sediment. Is there another possibility?

2. In general, is it possible to add hydraulic discontinuities (dam, weir, bridges) ? They might solve my first problem.

3. This weir is built at the junction of the river after a large island. I know that the Mohid River Network cannot cope with islands. But as an item on my wishlist: Is it very hard to add the possibility of separating the river by an island?

Thank you for all,

Cyril

[davidbrito]

Hello Cyril,

1. For now the erosion deposition parameters (e.g. critical shears) are not space dependent as in runoff. That is something that we will add to the TO DO list. The variables would have to go from scalar to matrix and be read in the beggining (if user specify something in nodes it is used, if not used the default vaule).

2. There is this option for discharges. Yo can have a discharge that is a weir and it will remove water (I think)if water goes higher the level of weir that you define.
http://wiki.mohid.com/wiki/index.php?title=Module_Discharges
I have never used it in MOHID Land.

3. I do not remember now why this was a problem. I have a unit test with only drainage network and it goes around an "island" and water is able to go down and up. Maybe the problem is the automatic delineation. You will have to remind me what the problem is that it does not allow sorry.


Best regards,
David

[cyrilgarneau@gmail.com]

Hi David,

Thank you for this complete answer.

1. I will have to manage the erosion problem in another way.

2. Actually, I do not wish to remove water from the river. I simply experience some numerical instability around weirs and under bridges. But still, I believe that specific handling of these constructions could reduce instability and prevent the use of Courant number (which multiplied by 5 or so the computation times). This could also solve the erosion problem from point 1. But at this moment, it is more of a wish than a need.

3. Could you provide a very simple drainage network example (4-5 nodes and reaches) ? I built my river network manually, so I will have to modify it accordingly.

Thank you again.
Cyril

[cyrilgarneau@gmail.com]

David,

I just tried to create an island in my model by adding reaches and nodes to the drainage network. Unfortunately, I get the following error:

Node with more then one Downstream reaches [ID] 12

Which means that MRN does not yet accept downstream separation of rivers.

Do you experience the same?

Thank you,

Cyril

[Jauch]

Hello Cyril,

At this moment, the ModuleDrainageNetwork do not accept multiple downstream reaches.
At the initialization of the module, the network is constructed and this condition is tested. Any node with more than one downstream reach will cause an error and the model will stop.

This verification is done at the subroutine ConnectNetwork.

Best regards,
Eduardo.

[davidbrito]

I have to verify if in the past there was not this limitation.

Because if you run the MOHID Studio MOHID Land Unit Tests there is one called "River 1D" were a river exists (run by MRN alone). I did not tested this simulation recently but it was working in the past you could see water go uo and downd the island.


Best regards,
David