Derivatives of U and V in curvilinear coordinates

[fcampos]

Hello
I want to compute the derivative of U and V in curvilinear coordinates but in sigma (no Z) with the aim of compute the conversion of barotropic energy (similar to Marchesiello et al 2003). I did the code in matlab but I'm sure it's good. Can you help me please? I used part of the ROMS-Agrif Visualizations_tool code

Thank you very much.

[lalvarez]

Hello fcampos

You have this
uom(i,:,:) = 2*squeeze(Hu(i,:,:))./(pm(:,1:L)+pm(:,2:Lp));
uon(i,:,:) = 2*squeeze(Hu(i,:,:))./(pn(:,1:L)+pn(:,2:Lp));
von(i,:,:) = 2*squeeze(Hv(i,:,:))./(pn(1:M,:)+pn(2:Mp,:));
vom(i,:,:) = 2*squeeze(Hv(i,:,:))./(pm(1:M,:)+pm(2:Mp,:));

how about
uom(i,:,:) = squeeze(2*((Hu(i,:,:))./(pm(:,1:L)+pm(:,2:Lp))));
uon(i,:,:) = squeeze(2*((Hu(i,:,:))./(pn(:,1:L)+pn(:,2:Lp))));
von(i,:,:) = squeeze(2*((Hv(i,:,:))./(pn(1:M,:)+pn(2:Mp,:))));
vom(i,:,:) = squeeze(2*((Hv(i,:,:))./(pm(1:M,:)+pm(2:Mp,:))));

perhaps it works

[fcampos]

Hello.
The 2 affects all the matrix. So if you put it before or after "squeeze" affect the multiplication in a similar way. The purpose of my question is if the structure of the transformation into sigma coordinates (in matlab) is indeed good. in my opinion it is well written XD.

Thank you for your reply