Welcome, Guest
Username: Password: Remember me
Forum on HARMONIE Surface development

TOPIC: Crazy clay/sand fileds for 1km res. domain

Crazy clay/sand fileds for 1km res. domain 3 years 6 months ago #1543

Hi!

Has recently discovered that cy40h1 (with SURFEX7.3) creates crazy sand/clay fields if the operational versions of clay/sand at 10 km are combined with a "very" high resolution domain (1 km in our case).

Attached pdf

File Attachment:

File Name: Sand_results.pdf
File Size: 3,179 KB
shows a domain over Stockholm region. Sorry for the low quality figures but the colour scale is the same in all figures showing percentage of sand between 0 and 100%.

Upper left figure: raw 10km sand field.
Upper right figure: sand in PGD file with standard settings. This creates a crazy stripe pattern of sand (which will influence e.g. soil moisture).

In standard setting the parameter NHALO is set =2. It can be changed in surfex_namelists.pm by
%pgd=(
 NAM_IO_OFFLINE=>{
  CSURF_FILETYPE => 'CSURF_FILETYPE,',
  CPGDFILE      => 'CPGDFILE,',
  NHALO          => '23,',    
 },

Lower left figure: A test with NHALO=5. Looks much better but some strange values still appear.
Lower right figure: Here NHALO=23 is used. This is the value one needs to use if the same exercise is performed in SURFEXv8 since v8 will abort with a message that NHALO has to be increased if it is too small. NHALO=23 is the lowest allowed NHALO in v8.

Ekaterina Kourzeneva at FMI has looked into this problem of interpolation with respect to lake depth and she advised me on how to solve the problem. Thanks Katya!!

So, in conclusion, be careful to check your PGD results when you combine low resolution physiography with high resolution domain setting.

Cheers,

Patrick
Time to create page: 0.092 seconds