[Thread Prev][Thread Next][Index]

Re: [ferret_users] Calculating relative vorticity



Hi Paulo Santiago and Savin,

| Supposing you are using relative vorticity defined as
| 
| zeta = du/dx - dv/dy
| 
| in Ferret, you can do something like:
| 
| let dudx = u[i=@ddc]
| let dvdy = v[j=@ddc]
| let zeta = dudx - dvdy

Except that zeta = dv/dx - du/dy  , assuming that u and x are
eastward and v and y are northward.

I have two more caveats:

1) The formula above is correct only when u and v are defined on the
  same grid;

2) The formula above is an approximation on a sphere.  The real curl
  is dv/dx - d(u cos(phi))/dy/cos(phi)  or something like that
  (please don't believe it without checking), where phi is the
  latitude, dx = a cos(phi) d(lambda), and dy = a d(phi).
  If you want to include the cosine factor you can use

     let c = cos(pi*y[gy=u]/180.0)
     let cu = c * u
     let dudy  = cu[j=@ddc]
     let dv/dx = v[i=@ddc]
     let zeta = dvdx - dudy/c

  or along the lines.

Cheers,
Ryo


[Thread Prev][Thread Next][Index]

Contact Us
Dept of Commerce / NOAA / OAR / PMEL / TMAP

Privacy Policy | Disclaimer | Accessibility Statement