Frequency and The Redshift Factor: Notes For GR-I - CCD

Frequency and the Redshift Factor
Notes for GR-I - CCD
Waves and Rays

We consider the propagation of electromagnetic radiation in a curved spacetime. We expect
to have to consider both the wave aspects and the ray aspects of this radiation.
Consider a family of hypersurfaces where each member is dened by the constancy of
the function S(x
c
) over that hypersurface. S(x
c
) can be interpreted as the phase function
of a propagating wave. A vector eld normal to the hypersurface S(x
c
) = C, where C is a
constant, is collinear with the vector eld n
a
= S
|a
, using our usual notation for derivatives.
Since we are particularly interested in the propagation of photons, we will take S(x
c
) = C
to be a null hypersurface, i.e. n
a
n
a
= 0, so that n
a
is a null vector eld. Since n
a
is the
gradient of a scalar function, this vector eld is curl-free, that is n
a||b
n
b||a
= 0.
Let be a member of the congruence of curves that pierce the hypersurface S(x
c
) = C
orthogonally, and suppose that k
a
is a tangent vector to , so that k
a
is collinear with n
a
.
Thus we will take k
a
= f(x
c
)n
a
, where f(x
c
) is some function to be determined.
To further determine the nature of the family of curves of which is a member, consider
the covariant derivative
k
a||b
k
b
= (fn
a
)
||b
fn
b
(1)
= fn
b
_
f
|b
n
a
+ fn
a||b
_
(2)
= f
|b
n
b
k
a
+ f
2
n
a||b
n
b
(3)
= f
|b
n
b
k
a
+ f
2
n
b||a
n
b
(4)
Since n
b
n
b
= 0, we have
0 = (n
b
n
b
)
||a
= n
b
||a
n
b
+ n
b
n
b||a
= 2n
b||a
n
b
(5)
so that we then have
k
a||b
k
b
= f
|b
n
b
k
a
(6)
so that k
a
is parallel transported along the curve . Thus is a null geodesic curve. If we
now choose f
|b
n
b
= 0, so that f is a constant along , we nally have k
a||b
k
b
= 0. Thus
with this adjustment, k
a
= dx
a
/d is the null tangent vector along with ane parameter
. This completes the picture of the propagation of null radiation, with the surfaces
c Charles C. Dyer
1
S(x
c
) = constant being the wavefronts corresponding to surfaces of uniform phase, and
being a typical ray, with tangent vector k
a
.
By the appropriate choice of the proportionality, we then have that k
a
= S
|a
measures
the rate of change of the phase at a xed point in the coordinate system, x
c
. Consider now
a time-like observer with 4-velocity u
a
= dx
a
/d, where is his proper, that is his own
wrist watch time. Then the rate of change of phase in his own proper time is
dS
d
= S
|a
dx
a
d
= k
a
u
a
(7)
It then follows that the frequency of the radiation as measured by this observer using his
own clock is proportional to u
a
k
a
. We normally choose to write this frequency with the
proportionality chosen such that
= 2u
a
k
a
(8)
Redshift from Maxwells Equations
We restrict consideration to a static spacetime, in the coordinate system where g
ab
is
independent of the time coordinate, x
0
= t, and g
0
= 0 where Greek indices indicate
spatial components.
Recall that the electric and magnetic elds can be represented by the Faraday tensor,
F
ab
. This tensor is antisymmetric, and can be obtained from A, the 4-potential, as the curl
F
ab
= A
b||a
A
a||b
= A
b|a
A
a|b
(9)
From the Maxwell equations, one can show that A satises the equation
_
gg
ma
g
nb
_
A
b|a
A
a|b
__
|n
=
g
c
j
m
(10)
where j
m
is the 4-current.
Since we are interested in the propagation of electromagnetic radiation of wavelength
much smaller than the radius of curvature of the underlying spacetime, we will take the
spatial part of the 4-potential to be of the form
A
(x
c
) = a
(x
)e
t
(11)
where a
(x
) is a slowly varying amplitude, a function of only the spatial position, and e

t
represents the wave behaviour with time, t being the phase of the wave. We take A
0
= 0,
and choose a gauge where
_
gg
00
A
_
|
= 0 (12)
which is related to the Lorenz gauge in 3-D, since
_
|
3
g| = g
00
g.
Upon substituting for A
b
in the wave equation, we obtain the time independent wave
equation for the amplitude a
g
_
gg
_
a
|
a
|
__
|
+

2
c
2
g
00
a
= 0 (13)
2
and we observe that the usual
2
factor in the last term is replaced by the term
2
g
00
, so
that the observed frequency is modied by the factor
_
|g
00
|.
Consider an emitter of radiation, P
1
, sending a photon to a receiver, P
2
. We will take
both to be at rest in the coordinate system (t, x
). Observer P
1
will measure his proper
time, , on his wrist watch, not the coordinate time, t. As P
1
moves along his world-line,
both and t change, and the relationship between them is
d =
g
00
dt (14)
The frequency that P
1
measures for the emitted radiation is simply the rate of change of the
phase of the electromagnetic wave as measured on his own wrist watch. Thus considering
the corresponding intervals dt and d, the measured frequency is
1

dt
d
=

g
00
=
_
g
00
all at P
1
(15)
This is consistent with the appearance of the term
2
g
00
as the position dependent angular
frequency in the time independent wave equation we have considered.
We can repeat the calculation to obtain the frequency that observer P
2
will measure,
nding that
2

dt
d
=

g
00
=
_
g
00
all at P
2
(16)
Dening the redshift z = /, where is the measured wavelength of the radiation, we
can dene the redshift factor to be
1 + z =

1
2
=
_
g
00
(2)
g
00
(1)
(17)
Our result is consistent with our previous result that the measured frequency is pro-
portional to u
a
k
a
, although the result we have just obtained is much more restricted, given
the additional assumptions we made to obtain it.
Redshift in the Schwarzschild Vacuum
Consider the case of the Schwarzschild vacuum solution, in the usual coordinates. Then
we have
2
=
_
1 2m/r
2
1 2m/r
1
(18)
where the source and observer are at Schwarzschild radial coordinates r
1
and r
2
respectively.
In the regime where both r
1
and r
2
are much larger than 2m, we can expand this, and retain
the lowest order terms to obtain
2
1 + m
_
1
r
1
1
r
2
_
(19)
Recalling the form of the gravitational potential in Newtonian gravity, we can write this
result in the form
2
1 +

2
1
c
2
(20)
3
Thus the change in frequency appears, in the classical approximation, to be just the energy
of the photon, h, that is expended in climbing out of the Newtonian gravitational potential
in going from r
1
to r
2
where for the moment we take r
2
> r
1
.
The Doppler Redshift
Consider two observers, A and B, located at the same point (or at least very close)
in spacetime. Suppose that A sends some photons to B, along the x-axis, say. Then the
4-velocity, k
a
, of the photon has components (1, 1, 0, 0), where we are setting the speed
of light, c, to one. Let
A
u
a
and
B
u
a
be the 4-velocities of A and B respectively. We will
assume that these two observers are only moving, spatially, along the x-axis, with speeds
A
v
x
and
B
v
x
, both taken to be very small when compared to the speed of light, so that both
A
v
x
and
B
v
x
are small relative to unity, and their squares can be dropped. The 4-velocities
of the particles are then
A
u
a
(1,
A
v
x
, 0, 0) and
B
u
a
(1,
B
v
x
, 0, 0) (21)
The redshift factor for the photon sent from A to B is then
1 + z =
A
u
a
k
a
B
u
a
k
a
=
1
A
v
x
1
B
v
x
1 (
A
v
x
B
v
x
) (22)
We thus recover the classical Doppler eect, since the redshift is just proportional to the
relative spatial velocity between the source and observer, where it is important to recall
that we have set the speed of light, c, to unity.
The Cosmological Redshift
time
space
k
a
A
u
a
B
u
a
World-line of A World-line of B
p
h
o
t
o
n
p
a
t
h
Figure 1: A photon sent from observer A to
observer B.
Turning now to the simple FLRW cos-
mological models, we consider a source A
and an observer B, each comoving with
the global spatial expansion. Thus the 4-
velocities are
A
u
a
= (1, 0, 0, 0) and
B
u
a
= (1, 0, 0, 0)
(23)
Solving the geodesic equations for a pho-
ton propagating in the FLRW models one
has the result that k
0
1/R(t), where
R(t) is the scale factor describing the expan-
sion or contraction of the homogeneous and
isotropic spatial sections. There is no need
to obtain the other components of the null
geodesic tangent vector k
a
for our present
purposes, since both the source and observer
are at spatial rest in these coordinates. We
can then compute the redshift factor be-
tween these two observers, A and B, to be
1 + z =
A
u
a
k
a
B
u
a
k
a
=
R(t
B
)
R(t
A
)
(24)
4

Frequency and The Redshift Factor: Notes For GR-I - CCD

Uploaded by

Copyright:

Available Formats

Frequency and The Redshift Factor: Notes For GR-I - CCD

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Frequency and The Redshift Factor: Notes For GR-I - CCD

Uploaded by

Copyright:

Available Formats

Frequency and the Redshift Factor

Notes for GR-I - CCD

Waves and Rays

) is a slowly varying amplitude, a function of only the spatial position, and e

You might also like