Is there any difference for different bed layer number?

[leon]

Dear all,
Recently, I'm doubtful about the number of sediemnt bed layers. You can set three bed layers or five bed layers or some else. Is there any difference? I read the paper Warner et al. wrote in 2008. What I got is that 3 bed layers maybe ok. So what's the effect of setting more bed layers? Are there any benifits?
Any suggestion is ok.
Thank you very much~

[jcwarner]

well, the number of bed layers sets the number of layers that compose the sediment bed at each rho point. The purpose of this is to allow the stratigraphy of the bed to evolve. if you are not interested in stratigraphy, then just set nbed = 1. Else, you can allow the bed to track changes. lets say for example you have river that delivers a pulse of sediment. this sed spreads out on the inner shelf. then another river pulse happens and another pulse of sediment deposits. If you have multiple bed layers, they can track how the bed is deposited on each pulse. There are many examples that anyone can think of, but in the end you have to decide if this method is what you need for your particular application.

[CBian]

Dear jcwarner,

I tried use Nbed=1,and Nbed=5 in my model, other configurations are same, but the sediment mass transport results are very different. Even use same Nbed=5, I tried different bed layer thickness, after 4 months, the mass transport results are also have obvious different. I use the latest version of ROMS. Is there something wrong in my model configuration?

[jcwarner]

there can be differences in the results depending on how you initialize the bed. If you set 5 layers, and most of the fine material gets eroded away, then that part of the bed will become 'armored' and the other sediment may not be able to be suspended. All of this depends on the particular application, stress distribution, etc. So look at your simulation, stress history, and the sediment layers and see if it makes sense.