[Mitgcm-support] ecco_c44_c23 major bug issues.

[mitgcm-support at dev.mitgcm.org]

The call to cost_sst is missing in cost_hyd. 

Cost final: 
it prints out objf_tauum twice.
Change objf_tauum to objf_tauvm.

Also, in the summation of cost contributions, the multiplier for
obj_tauvm should be mult_tauv, not mult_tauu.

Mult_hflux and mult_sflux aren't right, either.

Cost_weights:
Why did DS comment out the cosphi?
Regarding the factor 5 in the wsst calculation, why would DS want to
downweight the Reynolds SST w.r.t Levitus?

The weights wsalt2 and wtheta2 are "inverted" twice. This doesn't make
any sense at all.

On roughly line 896 of cost_weights.F , wtheta is set to 1. This should
be wtheta2 = 1.

When calculating wsflux and whflux, a magical factor of 3 has appeared
without any reason (to me). 

Jake