Community header

Forum

Questions regarding the real world case model

MV
Mathijs van Oostrum, modified 2 Years ago.

Questions regarding the real world case model

Keen Forecaster Posts: 2 Join Date: 4/12/19 Recent Posts

Hello forum users,

 

I'm currenly using the real world case model of the Riverlab for my bachelor assignment.

In the process I stumbled accross several things that are unclear to me.

- When I edit the BcFil in the Waal_gr_RE.mor file, I obtain the same results. Why does this happen, do any other changes ahve to be made?

- What does the MorFac do? I understand it is an acceleration factor, but by how much does it accelerate?

 

I hope somebody can help me out.

Thanks in advance!

Willem Ottevanger, modified 2 Years ago.

RE: Questions regarding the real world case model

Keen Forecaster Posts: 3 Join Date: 9/20/11 Recent Posts

Hi Matthijs, 

 

- When I edit the BcFil in the Waal_gr_RE.mor file, I obtain the same results. Why does this happen, do any other changes have to be made?
 

If I understand your question correctly, in the Waal_gr_RE.mor file you have changed the keyword 

    BcFil          = #Waal-1cm.bcm#

to 

    BcFil          = #Waal-2cm.bcm#

which is a file also available in the model directory.

Are there morphological changes yet or is it still in the spin up phase? In other words: has your model run long enough? The bed level point which is changed lives in the ghost cell, and is not available in the model output, so it is not possible to check this directly. If the result is exactly the same this needs to be looked into. 

 

- What does the MorFac do? I understand it is an acceleration factor, but by how much does it accelerate?

The changes to the bed level are increased by a multiplication factor every time step. Details about the morphological updating can be found here:

https://content.oss.deltares.nl/delft3d/manuals/D-Morphology_User_Manual.pdf#section.8.6

 

In 1D model the updating is a little bit more complex, because the bed level is the deepest point in the cross-section and the erosional/depositional volume is multiplied by the morphological factor. The redistribution of the erosion or deposition in the cross-section happens relative to the level at which the main channel width occurs. 

 

 

MV
Mathijs van Oostrum, modified 2 Years ago.

RE: Questions regarding the real world case model

Keen Forecaster Posts: 2 Join Date: 4/12/19 Recent Posts

Hi Willem,

 

Thanks for the quick response!

 

1) I did indeed change the keyword as you described and the file is available in the model directory. 

Morphological changes definitely occured. When plotting it is visible that the results are exactly the same at the same timestamp.

 

2) So if I'm correct, when running the model for a year, morphological changes for 10 years are simulated? However, the results obtained when running the model for a year appear as the results that are expected after 100 years. At some upstream locations over 1 meter of degradation can be observed, where at some downstream locations over 1 meter of aggredation can be observed (see attachment, the blue line corresponds with 31-jan-2000 and the red line corresponds with 31-jan-2001). Is there an explanation for this behavior? 

Willem Ottevanger, modified 2 Years ago.

RE: Questions regarding the real world case model

Keen Forecaster Posts: 3 Join Date: 9/20/11 Recent Posts

Mathijs van Oostrum:

1) I did indeed change the keyword as you described and the file is available in the model directory. 

Morphological changes definitely occured. When plotting it is visible that the results are exactly the same at the same timestamp.

2) So if I'm correct, when running the model for a year, morphological changes for 10 years are simulated? However, the results obtained when running the model for a year appear as the results that are expected after 100 years. At some upstream locations over 1 meter of degradation can be observed, where at some downstream locations over 1 meter of aggredation can be observed (see attachment, the blue line corresponds with 31-jan-2000 and the red line corresponds with 31-jan-2001). Is there an explanation for this behavior? 

1) Ok we probably need to look into this. Meanwhile you could try to prescribe other types of boundary conditions,

See the section after BcFil here: 

https://content.oss.deltares.nl/delft3d/manuals/D-Morphology_User_Manual.pdf#section.A.2

2) Yes, that is the assumption. You could do a sensitivity analysis to the MorFac to see if this indeed holds. As to the results you observe, the model has not been calibrated, so that may explain the difference you see.

   

JM
Jacob A A Morgan, modified 11 Months ago.

RE: Questions regarding the real world case model

Keen Forecaster Posts: 1 Join Date: 4/27/18 Recent Posts
<p>Regarding point 1, has this been resolved? We have been having a very similar problem. We have been testing a model aimed at exploring changes in sediment supply. We are using the .bcm file to specify the volumetric rate of bedload supply. However, when we look at results, the bedload at the upstream boundary is always the same regardless of the values in the .bcm file. It sounds the same as the problem here.</p>