Astron. Astrophys. 323, 295-304 (1997)
Appendix A: determination of
A.1. First-order model I
The first-order particle orbits are
![[EQUATION]](img164.gif)
where the perturbation potential is related to the initial
gravitational potential through the relation
![[EQUATION]](img165.gif)
The inversion of (A1) is then:
![[EQUATION]](img166.gif)
and therefore is given by:
![[EQUATION]](img167.gif)
Thus, we obtain:
![[EQUATION]](img168.gif)
![[EQUATION]](img169.gif)
From this we derive the well-known result for the Zel'dovich
approximation used in the Sachs-Wolfe effect that
![[EQUATION]](img170.gif)
since
![[EQUATION]](img171.gif)
A.2. Second-order model I
Eqs. (17)(18) in terms of the perturbation potentials read
![[EQUATION]](img172.gif)
where
![[EQUATION]](img173.gif)
![[EQUATION]](img174.gif)
![[EQUATION]](img175.gif)
![[EQUATION]](img176.gif)
We write the perturbation potentials in terms of
and by inverting
(A9):
![[EQUATION]](img177.gif)
where
![[EQUATION]](img178.gif)
![[EQUATION]](img179.gif)
![[EQUATION]](img180.gif)
![[EQUATION]](img181.gif)
Using (A14), we invert (5) in the form
![[EQUATION]](img182.gif)
where
![[EQUATION]](img183.gif)
![[EQUATION]](img184.gif)
Therefore, given the condition of parallelism(19), we have that
(A19) reduces to (20) with
![[EQUATION]](img185.gif)
![[EQUATION]](img186.gif)
A.3. Second-order model II
Following the same procedure as in Appendix A.2, we write
![[EQUATION]](img187.gif)
where
![[EQUATION]](img188.gif)
![[EQUATION]](img189.gif)
![[EQUATION]](img190.gif)
![[EQUATION]](img191.gif)
Therefore, inversion of (A24) leads to:
![[EQUATION]](img192.gif)
where
![[EQUATION]](img193.gif)
![[EQUATION]](img194.gif)
![[EQUATION]](img195.gif)
![[EQUATION]](img196.gif)
![[EQUATION]](img197.gif)
![[EQUATION]](img198.gif)
![[EQUATION]](img199.gif)
Appendix B: Jacobian
The Jacobian is given by:
![[EQUATION]](img201.gif)
![[EQUATION]](img202.gif)
where
![[EQUATION]](img203.gif)
![[EQUATION]](img204.gif)
![[EQUATION]](img205.gif)
![[EQUATION]](img206.gif)
![[EQUATION]](img207.gif)
![[EQUATION]](img208.gif)
![[EQUATION]](img209.gif)
![[EQUATION]](img210.gif)
and the functions and
are model-dependent. For instance, for the
case of model I of Sect. 3,
![[EQUATION]](img213.gif)
![[EQUATION]](img214.gif)
where and are
defined in Appendix A.2. The assumption of parallelism (19) adopted,
is simplified by making the following
substitutions in (B1):
![[EQUATION]](img217.gif)
where
![[EQUATION]](img218.gif)
Hence, all terms containing are set equal
to zero, and thus (B1) simply reads
![[EQUATION]](img220.gif)
© European Southern Observatory (ESO) 1997
Online publication: June 5, 1998
helpdesk.link@springer.de  |