[Mitgcm-support] Re: initialisation

[mitgcm-support at dev.mitgcm.org]

Patrick Heimbach wrote:
> Why are there two routines to initialise the model fields
> - ini_dynvars
> - ini_fields
> I don's see the sense, some fields are initialised twice.

J-M added it to initialize uniformly everything in DYNVARS.h

I agree some of it is redundant but these adjoint people
insist that everything is initialized properly. J-M has
taken it to heart. I think his feeling is that everything
should be set to zero and then later what ever shouldn't
be zero should be set there (later). Kinda makes sense
but it's a bit of a mish-mash since we set things to zero
later too.

The M and m in MITgcmUV stand for Mish and Mash.

A.