Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

How is nu_div (weighting of divergence component when applying hyperviscosity) used?

Hello All:In user_nl_cam, there is a parameter called "nu_div". It seems to me that CAM5 damps divergent flow using "nu_div" parameter, not "nu". Is this correct? Does anybody know exactly how "nu_div" and "nu" is used and how they differ?Thank you very much.
 

goldy

Meteorologisk institutt
Staff member
'nu_div' is the hyperviscosity applied to divergence component of winds [m^4/s]. Its use in CAM5 is documented in:

Dennis, J. M., J. Edwards, K. J. Evans, O. Guba, P. H. Lauritzen, A. A. Mirin, A. St-Cyr, M. A. Taylor, and P. H. Worley (2012), CAM-SE: A scalable spectral element dynamical core for the Community Atmosphere Model, Int. J. High. Perform. C., 26(1), 74–89, doi:10.1177/1094342011428142

Its default value is 1.5x10^16 for the ne16np4 grid, 2.5x10^13 for the ne120np4 grid, and 2.5x10^15 otherwise.
If you set the value of 'nu_div' to be less than zero, the value of 'nu' is used instead. The value of 'nu' (the hyperviscosity coefficient for u,v, T [m^4/s]) is dependent on the grid resolution.
 
Thank you, Goldy.I read Section 2.9 of that paper. Although it explained the use of "nu", it is still ambiguous how "nu_div" applies on Equation (17) and (18). There are two terms on RHS of Eq (18), which can be plugged into Eq (17) to form four terms in the big parenthesis on RHS of Eq (17). When you said "applied to divergence component of winds", are you referring to the 1st term on the RHS of Eq (18)?Thanks again. I appreciate your help. 
 

goldy

Meteorologisk institutt
Staff member
I can see why that is confusing. To really understand this, I strongly encourage you to contact the authors for their take.The way I think of it, is that they apply the ratio, "nu_div" / "nu", directly to the divergence (nabla cdot vec{u}) but I defer to the authors on that interpretation.
 
Top