Geoscience Reference
In-Depth Information
π
4 + nπ < Im ¯ k< 3 π
+
4
=0 . When u =0 , function w k has a root k = 0 which is not the
root of (8.38). Equation (8.A.3) results from (8.A.2) when changing k for
for n
k
Thus, roots of (8.36) do not exist.
Appendix M agnetic mode. F inite σ g
Equation (8.36) at σ g
=
has the form
k
exp 2 k + K
2
K k d g
k d g
K
2
=
κ g .
(8.B.1)
Equation (8.B.1) is obtained at Re k A
k 2 > 0. We investigate (8.B.1) at
Re κ g > 0, Im κ g > 0 . Assume that roots satisfy the condition
k d g |
|
1 ,
(8.B.2)
then
1
k d g 1
.
2 i
( k d g ) 2
i
( k d g ) 2
κ g = k d g
Substituting this expression for κ g into (8.B.1), we get
exp ξ
1
exp ξ
= K
,
(8.B.3)
ξ
where A = τ K d g / 2 =2 k . Equation (8.B.3) is solved by successive ap-
proximations with respect to parameter τ K with fixed A . Then, in the first
approximation,
exp ξ 0 = 0 . (8.B.4)
We restrict our consideration to the band 0 < Im ¯ k< 4 π .At A>e =2 . 71 this
equation has two real roots, the bigger one can be estimated by the formula
ξ 0
ln ( A (ln A (ln A... ))) .
(8.B.5)
Complex roots appear at A<e . We present (8.B.4) as two real equations for
Re ξ 0 = x 0 and Im ξ 0 = y 0 :
exp x 0 cos y 0
Ax 0 =0 ,
exp x 0 sin y 0
Ax 0 =0 .
(8.B.6)
The values of y 0 and x 0 can be found from
exp ( y 0 cot y 0 ) sin y 0
y 0
= A,
x 0 = y 0 cot y 0 .
 
Search WWH ::




Custom Search