[MITgcm-support] Fwd: Re: phiHydLow

[Martin Losch]

Sergey, Dimitris,
  phiHydLow is not the but *some* bottom pressure anomaly, scaled with a 
constant rho (rhoconst=999.8 by default, but that can be set in data, 
look up "your" value in the standard out, it's written to standard out 
by config_summary). To get the full bottom pressure, you'll have to add 
the the constant contribution of gravity*rhoConst*depth. If you choose 
a "clever" rhoConst in your data file then you can, in principle, get a 
zero mean.
The reason for all this is, as far as I remember:
the z-p ismorphism. The same variable, that is pressure (minus 
g*rhoConst*z) in z-coordinates (phiHyd) is geopotential height in 
pressure coordinates, so phiHyd is the "general" variable and for 
consistency, phiHydLow, which is the phiHyd at the k_bottom+1th 
interface is also computed in this way (it is gravity*sea surface 
height in pressure coordinates for the ocean). the part of 
gravity*rhoConst*z is substracted for numerical reasons (?): you 
substract a constant of order 1e4*z in order to increase relative 
precision.

for z-coordinates:  both sea surface height and (atmospheric) surface 
pressure are included in phyHydLow. So, in order to get real bottom 
pressure anomaly you have to
1. substract the horizontal mean (there is a subroutine REMOVE_MEAN_RL, 
currently commented our, that does this "online":  it's called from 
dynamics.F at the very end before #ifdef ALLOW_DIAGNOSTICS. You'll find 
a copy of the routine in 
verication/global_ocean_pressure/code/remove_mean_rl.F, put it in your 
"code" directory and comment in the call in dynamics.F, but your can do 
this "offline" as well)
2. scale by rhoConst

Hope that helps,

Martin


On Mar 5, 2005, at 7:14 AM, Dimitris Menemenlis wrote:

> Hi Sergey, it took me a while to figure this out too.
> phiHydLow is the depth integral of
> (rho - rhoconst) * g * dz / rhoconst
> Units are m^2/s^2.  The reason the value is not
> very small is that, unless you specify a different
> value in the data file, the default rhoconst = 999.8
> Don't ask me why, I think it's just historical
> and backward compatibility reasons?
> So the values you are seeing are roughly
> .3 * depth and would be pretty pretty meaningless,
> to you unless you remove a time mean.  You can convert
> to units of m by dividing by g = 9.8.
>
> phiHydLow definitely includes sea-surface height,
> and if you have atmospheric forcing turned on, it
> should also include atmospheric pressure ... but
> I'm just starting to get familiar with this part
> of the code so maybe someone with more experience
> can comment.
>
> Cheers, D.
>
> -- 
> Dimitris Menemenlis <menemenlis at jpl.nasa.gov>
> Jet Propulsion Lab, California Institute of Technology
> MS 300-323, 4800 Oak Grove Dr, Pasadena CA 91109-8099
> tel: 818-354-1656;  fax: 818-393-6720
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://dev.mitgcm.org/mailman/listinfo/mitgcm-support