Bed armoring

Sediment modeling collaborators: issues, applications, information exchange

Moderators: arango, robertson, rsignell

Post Reply
Message
Author
User avatar
stevenmiguelfigueroa
Posts: 22
Joined: Mon Apr 20, 2020 12:49 pm
Location: Chungnam National University

Bed armoring

#1 Unread post by stevenmiguelfigueroa »

Is bed armoring always implemented when using the sediment module?
Can bed armoring be turned on/off, or its properties changed?
Is bed armoring implemented through the active layer thickness?
Does it depend on the bottom stress algorithm used?

It does not seem to be mentioned directly in Warner et al. (2008), WikiROMS, or this forum.

In my application, I am starting with an estuary with 3 grain sizes well-mixed in a 100 m thick bed.
I run with SEDIMENT, SUSPLOAD, SED_DENS, SED_MORPH, NONCOHESIVE_BED1, UV_LOGDRAG.
I am not certain whether bed armoring is included, but I think it affects the result.

I appreciate any comment. Thank you.

jcwarner
Posts: 1172
Joined: Wed Dec 31, 2003 6:16 pm
Location: USGS, USA

Re: Bed armoring

#2 Unread post by jcwarner »

i am not sure what NONCOHESIVE_BED1 is,
but for bed armoring- there is not an explicit armoring routine. Instead, as the finer grain material is eroded, the surface will become composed of coarser grain sizes. this coarsening will increase the bottom(i,j,itauc), and thus create a higher critical stress for erosion, and reducing the amount of sediment that is transported.
if you want to change all that, it is in the sed_bed routines and the sed_surface routine.
you could add something that more directly computes itauc, in sed_surface.

-j

User avatar
stevenmiguelfigueroa
Posts: 22
Joined: Mon Apr 20, 2020 12:49 pm
Location: Chungnam National University

Re: Bed armoring

#3 Unread post by stevenmiguelfigueroa »

Dr. Warner,

Thank you for your response.

NONCOHESIVE_BED1 is an option which I didn't explicitly set in the .h file, but it appears in the log file in the list of options used.

So I gather that bed armoring is always implemented (albeit not explicitly) in the sediment module. I think it works through hydrodynamic sorting in the active layer. Different bottom stress algorithms compute different bottom skin friction stresses, and therefore active layer thicknesses (through the active layer thickness coefficient, k1). But neglecting that, bed armoring occurs regardless of the bottom stress algorithm used. And to turn bed armoring on/off, or change it's properties, it can be done through the sed_bed and sed_surface routines.

Does the number of bed layers and the bed layer thicknesses affect the bed armoring behavior? For example, consider a 100 m thick bed with four 25 m thick bed layers. If the active layer thickness is 1 cm, will the hydrodynamic sorting in the active layer ('bed armoring' effect) be distributed across the 25 m thick surface layer? If so, then I think the bed armoring does not work well if the top layer of the bed is very thick.

jcwarner
Posts: 1172
Joined: Wed Dec 31, 2003 6:16 pm
Location: USGS, USA

Re: Bed armoring

#4 Unread post by jcwarner »

yes the armoring effect would not work well if the top layer is very thick.
if you are interested in having multiple layers you may need to run some test simulations to see what would work best for you.
there is a flag 'new bed layer thick' set in the sediment.in that tells the model when to combine layers (from deposition) and it will split layers when the top layer is eroded.
layer thickness all depends on how active the sediment is.

User avatar
stevenmiguelfigueroa
Posts: 22
Joined: Mon Apr 20, 2020 12:49 pm
Location: Chungnam National University

Re: Bed armoring

#5 Unread post by stevenmiguelfigueroa »

Thank you for the easy to understand explanation.

Post Reply