[MITgcm-devel] seaice adjoint and EVP

[Dimitris Menemenlis]

Martin,

> I have corrected the bug in seaice_ocean_stress (I am wondering if this will
> fix the problem that Dimitris has with evp. I don't think so, but there is
> always hope o:), and updated lab_sea but not the

Unfortunately the stripes are still there.  This is not with very latest code. 
It is yesterday's pkg/seaice code with

> columbia15: diff seaice_ocean_stress.F ../MITgcm/pkg/seaice/seaice_ocean_stress.F 
> 184c184
> <           fuIceLoc=HALF*( DWATN(I,J,bi,bj)+DWATN(I-1,J,bi,bj) )*
> ---
>>           fuIceLoc=HALF*( DWATN(I,J,bi,bj)+DWATN(I,J+1,bi,bj) )*
> 195c195
> <           fvIceLoc=HALF*( DWATN(I,J,bi,bj)+DWATN(I,J-1,bi,bj) )*
> ---
>>           fvIceLoc=HALF*( DWATN(I,J,bi,bj)+DWATN(I+1,J,bi,bj) )*

So I am assuming EVP will eventually crash as before.
Figure comparing evp to lsr solution is here
http://ecco2.jpl.nasa.gov/data1/arctic/output/tests/evp/evp_vs_lsr.ps.gz
Directory
http://ecco2.jpl.nasa.gov/data1/arctic/output/tests/evp/
contains pkg/seaice output, including sigmas, matlab routine to read/plot and an 
additional figure
http://ecco2.jpl.nasa.gov/data1/arctic/output/tests/evp/evp.ps.gz
of all the evp fields
sigma* colorscale is chosen to emphasize open water.
character of sigma1 over open water is much different from sigma2 and sigma12, 
is this what you expect?

D.