Bug fix for COARE3.5 (not used by default) and move some parameters t… #143
Conversation
…o namelist options
| @@ -2418,7 +2419,7 @@ SUBROUTINE edson_etal_2013(Z_0,ustar,wsp10,visc,zu) | |||
| REAL, INTENT(IN) :: ustar, visc, wsp10, zu | |||
| REAL, INTENT(OUT) :: Z_0 | |||
| REAL, PARAMETER :: G=9.81 | |||
| REAL, PARAMETER :: m=0.017, b=-0.005 | |||
| REAL, PARAMETER :: m=0.0017, b=-0.005 | |||
There was a problem hiding this comment.
Is this change the reason the regression tests failed?
There was a problem hiding this comment.
I don't see how the code could reach this point since COARE_OPT is hard-coded to 3.0, so this subroutine is never called.
There was a problem hiding this comment.
Yes, it's currently hard-coded to the COARE 3.0 algorithm, so this isn't the reason for the failure. This change does make COARE 3.5 competitive with COARE 3.0. Beforehand, COARE 3.5 has a large negative 10-m wind speed bias:
|
These changes are in NCAR#911 and will be merged back to gsl/develop after the community repositories are updated. |
|
#143 will merge these changes |
| TH1D(I)=T1D(I)*THCON(I) !(Theta, K) | ||
| TC1D(I)=T1D(I)-273.15 !(T, Celsius) | ||
| TH1D(I)=T1D(I)**(100000./P1D(I))**ROVCP !(Theta, K) | ||
| TC1D(I)=T1D(I)-273.15 !(T, Celsius) |
There was a problem hiding this comment.
@joeolson42 Joe, this PR got the "buggy" MYNN surface layer scheme in. It looks like it did not get into the RRFS_dev branch
Two changes: