Scribd
Upload
689 views

Calculating Arakawa Jacobian

The Arakawa Jacobian scheme is a finite-difference approximation to the Jacobian operator that conserves both kinetic energy and enstrophy in the vorticity equation. It uses a 5-point or 9-point grid to calculate the Jacobian terms through central differencing. Arakawa showed that taking the average of the three Jacobian formulations conserves both enstrophy and kinetic energy, which single formulations do not. This scheme prevents non-physical instabilities in numerical weather prediction models.

Uploaded by

mak112005gmail
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
689 views

Calculating Arakawa Jacobian

The Arakawa Jacobian scheme is a finite-difference approximation to the Jacobian operator that conserves both kinetic energy and enstrophy in the vorticity equation. It uses a 5-point or 9-point grid to calculate the Jacobian terms through central differencing. Arakawa showed that taking the average of the three Jacobian formulations conserves both enstrophy and kinetic energy, which single formulations do not. This scheme prevents non-physical instabilities in numerical weather prediction models.

Uploaded by

mak112005gmail
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Arakawa Jacobian scheme

Introduction:
It is known that in the vorticity equation in the absence of horizontal divergence and vertical motion, the absolute vorticity of a parcel is conserved, d 1 dt ( + f ) = 0 Where, is the relative vorticity, and f the Coriolis parameter. The above equation can also be written as t ( + f ) + V . ( + f ) = 0 This equation is known as non-divergent barotropic equation. Where, V is the non divergent wind. If is the velocity stream function then it is known that = 2. Thus the equation 1 becomes 2 2 t ( + f ) + V . ( + f ) = 0 2 ( + f ) = V . ( 2 + f ) t

t (2 ) = V . ( 2 + f ) 2 t = - J( ,2 + f ) 2 t = J(2 + f , ) 2 = J(, ) t Where J is the Jacobian of and 2

1
M A Kulkarni | M Tech 2nd Semester Practical Notes

Jacobian:
Jacobian means matrix of all first order partial derivatives of a functions (it may be scalar or vector functions) with respect to another vector. Thus, Jacobian J for two functions A and B can be defined as follows.

J(A,B) =

A x B x

A y

B y
3

A B A B x y - y x

Its very easy to show that B B J(A,B) = xA y - y A x A A J(A,B) = yB x - x B y 4 5

Arakawa Jacobian:
The Jacobian in equation 2 should give unique value. In other words it should conserve the kinetic energy (V2) and the enstrophy (2). But the problem arises if one uses different approaches given in equations 3, 4 and 5. All of the three approaches give different answer to the Jacobian. The Arakawa Jacobian scheme is a finitedifference approximation to the Jacobian operator that conserves both kinetic energy and enstrophy. ( i, j+1) ( i-1, j+1) ( i+1, j+1) The grid used for Arakawa 5 and 9 points scheme is 2 6 5 given in figure 1. ( i-1, j) 3
(i,j)

( i+1, j ) 1

Arakawa suggested to use Central difference scheme all the three approaches mentioned above. Different combinations of the three Jacobians statisy the criteria of conservation of kinetic energy and the enstrophy.

7 ( i-1, j-1)

4 ( i , j-1)

8 ( i+1, j-1)
2
M A Kulkarni | M Tech 2nd Semester Practical Notes

Figure1 Representation of grid

The method suggested is given below. Different notations can be used to define the Jacobian. Similar to equation 3: J++ = x y - y x Lets, take x = y = d 6

1 J++ = 4d2 ( (1- 3) (2 - 4 ) - (4 - 4) (1 - 3 ) ) Similar calculations for equation 4 and 5 yield 1 J+x = 4d2 [ 1 (5 - 8 ) - 3 (6 - 7 ) - 2 (5 - 6 ) + 4 (8 - 7 ) ] 1 Jx+ = 4d2 [2 (5 - 6 ) - 4 (8 - 7 ) - 1 (5 - 8) + 3 (6 - 7 ) ]

7 8

It can be seen from the equation6 66 66 that it requires 5 grid points thus it is 5 6 while equations 7 and 8 requre 9 grid points. J++ is commonly used for short range NWP. But, it doesnt conserve either the kinetic energy or 2 therefore, it can give rise to non linear instability 1 if the integration process is prolonged. Whereas, J+x and 2( J++ + Jx+ ) conserve 2 and Jx+ and 1 ++ +x 2 2( J + J ) conserve V . 1 Arakawa and Lam (1981) showed that, J = 3 (J++ + J+x + Jx+ ) conserves both enstrophy and mean kinetic energy.

Reference: Arakawa A and Lam V R (1981), A potential enstrophy and energy conservation scheme for shallow water equation, Monthly weather review, 109, pp 18-36

3
M A Kulkarni | M Tech 2nd Semester Practical Notes

You might also like