ndown run with *NO* boundary value nudging?

Discuss the nesting capability within the model itself and any problems you might have run into.

ndown run with *NO* boundary value nudging?

Postby imfergus » Mon Apr 19, 2010 5:11 pm

Hello all,

I am running a modified version of WRF 3.0 that is coupled to ParFlow, a 3D variably-saturated groundwater model with integrated overland flow. ParFlow requires fairly high spatial resolution (~1km or finer) and the modified code, PF.WRF, isn't set up for nesting. In order to run real data cases, I am first using WRF to downscale NARR to 1km, then using this output and ndown to carry out subsequent PF.WRF runs at 1km resolution.

The interior 1km nest of our WRF downscaling run is 57 x 42 cells (nx x ny), and our PF.WRF domain is 45 x 32 cells. The PF.WRF grid perfectly aligned with the WRF 1km grid, only smaller. Given the small size of the PF.WRF grid and the perfect alignment between WRF and PF.WRF grids, we want to the PF.WRF domain to evolve freely -- i.e., WITHOUT boundary value nudging. I have two questions...

First, is it possible to completely eliminate boundary nudging -- e.g., set relax_zone to zero -- or does this cause problems?

Second, are significant boundary effects likely despite perfect alignment between the parent and child grids (i.e., the 57x42 WRF grid and the 45x32 PF.WRF grid)? We think this is the best available approach, but I am wondering if someone else sees a potential problem that we have overlooked?

Thanks in advance,
Ian
imfergus
 
Posts: 2
Joined: Mon Apr 19, 2010 4:57 pm

Return to Nesting

Who is online

Users browsing this forum: No registered users and 2 guests

cron