calculating terms in momentum equations

[Jay]

Hi,

I now have some GETM output to analyze. At each spatial point, I would like to calculate each of the vairous terms in the momentum equations as shown below, in order to investigate the relative contribuition of them to the acceleration. This will involve the calculation of velocity gradients along x, y, and z.

Should I first transform the data from sigma layers to equidistant zax (using gvc2zax?), and then calculate the gradients along x, y, and z? Or, in another way, first calculate the gradients along the native GETM grid, and then transform them to zax? This is especially confusing for me when I try to calculate the mixing terms in the governing equations, because the horizontal mixing coefficient Am should be prescribed to be along the coordinate lines (i.e. sigma layers) instead of geopotential lines.

I would really appreciate for any advices. Thanks.

Jay

[Jay]

Update: 

I have found the compiling option _MOMENTUM_TERMS_ to output those terms.

However, different from what I was expecting, tdv_u + adv_u + vsd_u + hsd_u + cor_u + epg_u + ipg_u does not close to 0.

Am I correct that the budget of these terms should always close? If not, in what aspects can I modify my getm run for the closure of the momentum equation?

Thanks!

Jay