User Forum of Software BASEMENT

Yang He

User

Registered: 2024-06-02

Posts: 8

Question about suspended load modelling

Hello,

I am currently working on modeling suspended load transportation in a 56 km river, which has 3,065 cross-sections with a 20 m gap between each. The river has four tributaries. I set it to be fixed bed. However, I am seeking clarification regarding the "Suspended load" section of the model.

The simulation I am running is based on hydraulic steady-state results. I initially set an SSC of 0 g/L across the entire river, representing clear water. For the upper boundary condition and the four tributaries, I set a suspended sediment input of 0.7 g/L. Since the unit for the upper boundary condition is m³/m³, I converted 0.7 g/L to 0.000264 m³/m³, using the sediment density of 2,650 kg/m³. The external source term for tributaries is in m³/s. For example, with a constant discharge of 7 m³/s for the first tributary, I calculated the SSC input as 0.001849 m³/s ((0.7 g/L * 7m^3/s) / (2650 kg/m³)). This approach was applied to all tributaries and run the model until steady state of SSC.

Given this setup, I expected the results to show a uniform 0.7 g/L concentration throughout the river (0.000264 m³/m³ at each cross-section). However, the results differ, with only the section between the upper boundary and the first tributary showing 0.7 g/L. The rest of the river exhibits a stair-step pattern, with SSC increasing after each tributary.

Additionally, I tried setting 0.7 g/L as the upper boundary condition in clear water with 0g/L for the four tributaries, which resulted in a 0.7 g/L distribution throughout the entire river. I expected some dilution of SSC along the river. But I observed that, before reaching a steady state, the SSC is not diluted by the additional discharge from any of the tributary.

The results differ significantly from my expectations. Could there be an issue with my expectations, or is there a potential flaw in my model?

Below is the relevant portion of my model for your reference:

"SUSPENDED_LOAD": {
              "BOUNDARY": [
                {
                  "string": "upstream",
                  "type": "suspension_discharge",
                  "file": "ssc_initial.txt",
                  "mixture": "unique"
                },
                {
                  "string": "downstream",
                  "type": "out_down"
                }
              ],
              "INITIAL": {
                "concentration": 0.0,
                "type": "global_value",
                "mixture": "unique"
              },
              "PARAMETER": {
                "scheme": "quickest",
                "sediment_exchange": "off",
                "diffusion_factor_dynamic": 1.0,
                "sink_velocity_type": "vanrijn",
                "local_slope": "off",
                "exchange_type": "concentrations",
                "pickup_factor": 0.0,
                "settling_velocities": [
                  0.152
                ]
              }

Thank you very much for your assistance.

Best regards,
Yang

Sudesh

User

Registered: 2021-07-13

Posts: 1

Re: Question about suspended load modelling

Hello Yang,

Can you please provide the model setup files in a zip folder? You can either email them to me or provide a link for download. Then I will have a look for any inconsistencies.

Yang He

User

Registered: 2024-06-02

Posts: 8

Re: Question about suspended load modelling

Hi!

I have prepared the model setup files and shared them with you via Polybox to your email.
(polybox link -> https://polybox.ethz.ch/index.php/apps/ … 798516379)

The folder “BASEMENT_model_problem” contains two cases:

Case 1: 0.7 g/L SSC for all tributaries.
Case 2: 0.7 g/L SSC from the upper boundary and 0 g/L SSC for the tributaries.

Both models have been run, and the results are kept for direct review.

Besides, the problem I posted here (https://forum.basement.ethz.ch/viewtopic.php?id=5537) was using the same model. In case 1, when I changed the “sediment_exchange” from off to on, the warnings and error will appear. When you look at the model, I would greatly appreciate any suggestions to help resolve this error also.

Thanks for your time and attention. I look forward to any suggestions.

Best regards,
Yang

Yang He

User

Registered: 2024-06-02

Posts: 8

Re: Question about suspended load modelling

Hi!

I tried a little bit more on the previously mentioned issues, specifically by modifying the test case “ST_BC_2a” on the website. During my testing, I found that the problems disappear when using the “mdpm” scheme. When using other schemes such as “quickest,” “quick,” or “upwind,” the same issues show again. Additionally, when using the “holly” scheme, the simulation will stop automatically at the beginning without any error information.

Attached the link is the modifed cases I used to test. There are two cases inside.
    https://ethz-my.sharepoint.com/:f:/g/pe … A?e=jjrelF

For both cases I extend the total running time.

In “ST-BC-2-a_modified_case1”, I directly added one tributary at “Qp40” with constant discharge 25 m^3/s to assess the possible dilution after the additional water input.

In “ST-BC-2-a_modified_case2”, the tirbutary “Qp40” was further set with a constant SSC, which is the same as upper boundary input, to see how the downstream SSC changes after the confluence of the tributary with the main river.

Only when I set the scheme “mdpm”, the upstream SSC will show the results I expected, which is dilution in case 1 and no SSC changes in case 2.

In both cases, only the “mdpm” scheme produces the expected results upstream, showing dilution in Case 1 and no SSC changes in Case 2. I am currently reviewing the formulas used in each scheme. Could this be due to strong diffusion or some other intrinsic property of the formula?

As my model needs to simulate sediment exchange, switching to “mdpm” isn’t an option. The “quickest” scheme is necessary to me.

Any information or suggestion would be greatly appreciated.

Best regards,
Yang