[MITgcm-support] Diagnostic calculation of buoyancy frequency with nonlinear equation of state

[Sonya Legg - NOAA Affiliate]

Hi all,

I'd like to hear about your experiences diagnosing the buoyancy frequency
in simulations with a nonlinear equation of state. Here's what I am doing:

- I'm using the MDJWF eos
- I'm assuming that the pressure used in the equation of state is simply
given by the depth, i.e. pressure(k) = -Z(k)
- I'm using centered differencing, i.e. N2 =
-g/rho0*(rho(k)-rho(k+1))/(Z(k) -Z(k+1)), where k increases downward
following MITgcm convention. Z=RC
- I calculate the density of both the upper and lower parcel if they were
at the pressure corresponding to the intermediate location, i.e. at (Z(k) +
Z(k+1))/2
-  This mostly produces sensible looking results. However, the particular
simulation that I am analyzing has some deep layers of almost homogeneous
fluid (remnants of earlier mixing), which are giving N2< 0 results (where I
would expect N2=0), despite no indication of any convective velocities in
those regions. (In these nonhydrostatic simulations there are other regions
with overturns and plenty of turbulence).

I'm thinking there must be something I'm missing, having grown up in
linear-eos world.
Anyone have a matlab N2 routine for nonlinear eos I can compare with?

Thanks
Sonya
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20220908/4573cec1/attachment.html>