Relativistic Doppler effect

Masatsugu Sei Suzuki

Department of Physics, SUNY at Binghamton
(Date: January 13, 2012)

1. Doppler effect
The Doppler effect is the apparent change in frequency (or wavelength) that occurs
because of motion of the source or observer of a wave. When the motion of the source or
the observer is toward the other, the frequency appears to increase. When the motion of the
source or observer is away from the other, the frequency appears to decrease

(v  us )t (v  us )t v  us
  
N f 0 t f0

(ur  v)t
ft 

(ur  v)  v  ur 
f     f 0
  v  us 

1
us (>0) is the velocity of sender approaching the receiver.
ur (>0) is the velocity of the receiver approaching the sender.
f0 is the frequency of the sender and f is the frequency of the receiver.

Fig. A receiver is stationary and a source is moving toward the receiver at the velocity vs.
v is the velocity of sound. vs <v.

________________________________________________________________________
2. Relativistic Doppler effect

2
Fig.2
ct
ct'

x'
B A

q
q
x

Fig.3 Doppler effect

We consider a moving light source that flashes with frequency f0. The observer at rest in
the S frame receives the light signal with a different frequency f (this is called the Doppler
effect of light). First we calculate the distances in the S' frame;

3
 OC  S'  (OB  BC ) S '  (OA) S ' cos  ( AB) S ' tan

4
v
1
 ct ' cos  ct ' sin   ct ' c
v2
1 2
c

Using the scaling factor

v2
1 2
OC  S  c OC
v2
  S'
1 2
c

we have

v2 v2 v v
1 2 1 2 1 1
OC  S  ct  c OC
v2
  S'  c ct '
v2
c 
2
c ct '
v
1 2 1 2 v
1 2 1
c c c c

The frequency of the light observed at the S frame is related to that of the light emitted in
the S' frame is

v2 v2 v v
1 1 1 1
OC  S  ct  c 2 OC
v2
  S' 
v2
2
c ct ' c 
2
c ct '
v
1 2 1 2 v
1 2 1
c c c c

((Formula of the Doppler effect))

v
1
f ct ' 1 c <1
  
f 0 ct  1
v
c

t  t '

4
with

v
1
 c 1
v
1
c

3. Derivation of relativistic Doppler effect II

Fig.4 Derivation of the Doppler effect

The light source moves at the velocity v along the x axis. The light is emitted from the light
source. The angle between the direction of the light and the x axis is . We assume that the
angular frequency for the observer in the S frame is  and that the angular frequency for
the light source in the S' frame is ' = 0. Then we get the relation

E v
p
E' c c

c v 2
1 2
c

where E   , E '   '  0 ,and px  k cos  .

5
  v  v
 k cos    cos  2
 '
 c c  c c
c v 2
v 2
1 2 1 2
c c

or

v
 (1  cos  )
' c
v2
1
c2

or

v2 v2
1 1
  ' c2  0 c2
v v
1  cos  1  cos 
c c

4. Consideration
We consider the physics meaning of the above equation using a figure below.

6
 v
c
v v
d light b
light light
a=p
v light O light v
e a
light light
light
v p v
f a= h
2v
g

Fig. Observer at the origin. The star moves at the velocity v along the x axis. The light is
emitted from the star.

(a) When  =  (longitudinal Doppler effect)

v2 v
1 2 1
  0 c  c 
0 0
v v
1 1
c c

(receding)

The observed wavelength becomes longer (red shift).

(e) When  = 0 (longitudinal Doppler effect)

7
 v2 v v v
1 2
(1  )(1  ) 1
  ' c  0 c c  0 c 
0
v v v
1 (1  ) 2 1
c c c

(approaching)

The observed wavelength becomes shorter (blue-shift).

(c) When  = 3 (transverse Doppler effect), we have

v2 v2
  ' 1    0 1 
c2 c2

(g) When  =  (transverse Doppler effect), we have

v2 v2
  ' 1    0 1   0
c2 c2

(b) When  = 5, we have

v2 v2
1 1 
  ' c2  0 c2
v 5 v
1  cos( ) 1
c 4 c 2

(f) When  = , we have

v2 v2
1 1
  ' c2  0 c2
v  v
1  cos 1
c 4 c 2

8
 w
w0

4
e e
v
b= =0.90
c
3

f d
1
g c
h a b
a
p p 3p 5p 3p 7p
0 p 2p
4 2 4 4 2 4

 1  2
Fig. Plot of  with  = v/c = 0.90. The green line denotes  = 0.
0 1   cos 
a longitudinal Doppler effect (receding)
e longitudinal Doppler effect (approaching)
c and g transverse Doppler effect.

________________________________________________________________________
5. Doppler effect from the Minkowski space-time diargam

We suppose that a source is located at the origin of the reference frame S, and that an
observer moves relative to the frame S at the velocity v, so that the observer is at rest in the
frame S'. Each emitted pulse travels with speed c. Suppose a first pulse is sent out at t = 0
when the observer is at the position x = x0, and suppose the (n+1)-th pulse is sent out at t =

9
n. This will have covered n periods of vibration, so that the measured frequency of the
1
source in S is   .


ct ct'
D

A x'
B
C

q
q x
O B
Fig.5

In the above figure, we assume that

v
tan   .
c

OA  CD 
S' S'  n '

OB   OA
S' S' cos , AB   EB   OA
S' S' S' sin 

Then we have

10
 OE S  
 k OE S'  
 k OA S ' (cos  sin  )
v v2 v v
1 1 1 1
 n ' c c 2  n ' c  n ' c  n
v2 v2 v2 v
1 2 1 2 1 2 1
c c c c

Then we have

v
1
 ' c
v
1
c

or

v v
1 1
1 1 c  f c  f
f ' 
 '  1 v
1
v
c c

The wave lengths in the S and S' are defined by

  c , and  '  c '

Then the ratio of the wavelength is given by

v
1
 ' c ' c 1
  (red shift)
 c 1
v
c
_________________________________________________________________________
6. Twin paradox
There are twin: Mary (the Moving twin) and Frank (the Fixed twin). We imagine that
each person sends equally spaced time signals of their own proper times to the other, The
cumulative counts of time signals for the whole trip are then compared. Suppose that each
person is transmitting f pulses per unit time. As Mary travels away from Frank, each
observer will receive the other's signal at the reduced rate

1 
f ' f
1 

11
But for how long? Here is the asymmetry. As soon as Mary reverses, she begins to receive
signals from Frank at the enhanced rate

1 
f " f
1 

With Frank it is quite different. The last signal sent by Mary before she reverses does not
reach Frank until a time L/c later. Thus for much more than one-half the total time Frank is
recording the Mary's signals at the lower rate f'. Only in the latter stages does Frank receive
pulses at the higher rate f".
Note that each observer receives as many signals as the other sends between start and
finish of trip. Frank is able to infer from his observations that it took place at the
midmoment of the journey time as measured by Mary, since equal numbers of signal
received by Frank at the two different rates f' and f" (See French, Special relativity).

12
 Bct

ct'

C A

x'

q
q x
O
Fig.7

13
In the S frame,

(CA) S  L , (OC ) S  vT

In the S' frame,

v
cT '
(OA) S '  cT ' , CAS'  (CE ) S '  (CF ) S '  cT ' sin   c
v2
1 2
c

v v2 v
cT ' 1 cT '
  c  cT v  vT
2
L  CA S  k (CA) S '  c c 
v2 v 2
v2 c
1 2 1 2 1 2
c c c

OC  S'  (OA) S ' cos 

cT '
v2
1
c2

v2
1
OC  S  cT  k (OC ) S ' 
cT ' c 2  cT '
v2 v2 v2
1 1 2 1 2
c2 c c

or

cT '
cT  >cT'.
v2
1 2
c

The time (in the S frame) of detecting one of twin arrives at the turnaround: TF

14
 cTF  OF   S  k (OF ) S '  k[(OF ) S '  (CF ) S ' ]
v v2
1 2cT '
cT ' c ) c
( 
2 2
v v v2
1 2 1 2 1 2
c c c
v
cT ' (1  )
 c
v2
1 2
c
v c
 cT (1  )  vT  cT  L(1  )
c v

or

L L
TF   .
v c

The time TE in the S frame:

L L
TE   .
v c

______________________________________________________________________
((Note))
In the above figure, we assume that

v
tan   .
c

OA S'  n '

OC   OA
S' S' cos , AC   EC   CF   CA
S' S' S' S' sin 

Then we have

15
 OE 
S  k OE   S'  
 k OA S ' (cos  sin  )
v v2 v v
1 1 1 1
 n ' c c 2  n ' c  n ' c  n
v2 v2 v2 v
1 2 1 2 1 2 1
c c c c

___________________________________________________________________
Similarly we have

v v
1 1
BF   OE 
S S  n 1 ' c  n '
2 1
c  n
v 1
v
1 2 1 
c c

v v
11
1 1
c  f c
f " 
1 1 ' 1  v 1
v
c c

_______________________________________________________________________
7. Red shift and Hubble’s law (special relativity)
We suppose that a source is located at the origin of the reference frame S. An observer
moves relative to S at velocity v. So that he is at rest in S’ (in Fig we use S1 instead of S’
for convenience). According to the special relativity, we obtain the Doppler effect for the
light as

v
1
 2
c   (1  v  v  ...)
' 
v
 (1  ) 1
v c 2c 2
c c

and

v
1
f ' f c
v
1
c

16
where c is the speed of light and f   ' f '  c . This means that a spectral line that
normally has a wavelength  is observed at a longer wavelength ’. Note that

1

v2
1
c2

The spectral line is shifted by an amount of    ' . The red shift of the galaxy (usually
noted by z) is given by

 ' v
z 
 c

(a) The red shift

The light from distant stars and more distant galaxies is not featureless, but has distinct
spectral features characteristic of the atoms in the gases around the stars. When these
spectra are examined, they are found to be shifted toward the red end of the spectrum. This
shift is apparently a Doppler shift and indicates that essentially all of the galaxies are
moving away from us. The measured red shifts are usually stated in terms of a z parameter.
The largest measured z values are associated with the quasars.

17
 Hydrogen red-shift example

(b) Hubble’s law

The relationship between the distances to galaxies and the red shift is one of the most
important astronomical discoveries of the twentieth century. This relation tells us that we
are living in an expanding universe. In 1929, Hubble published this discovery. According
to the Hubble’s law, the recessional velocity v of a galaxy is related to its distance r from
the Earth by

18
 v = H0 r,

where H0 is constant commonly called the Hubble constant and H0 = 75 km s-1 Mpc-1. Here
Mpc is a megaparsecs (parsec, 1 pc = 3.262 ly; light year, 1 ly = 9.462 x 1015 m = 63,240
AU).

pc = 3.262 x 9.462 x 1015 m = 3.0857 x 1016 m.

1 Mpc = 3.0857 x 1022 m.= 3.0857 x 1019 km

Since v = zc and v = H0r, then H0r = zc. Thus the distance to a galaxy is related to its red
shift by

zc
r
H0

(c) Big Bang

How long ago did the Big Bang take place?

r 1
T0  
v H0

where T0 is the same for all galaxies.

1 MPc  s 1 3.09  1019

T0 = 1/(75 km s-1 MPc-1)) =   year  1.3  1010 years=13 billion
75 km 75 3.156  10 7

years

Note that 1 year=365 x 24 x 60 x 60 = 3.156x107 sec. The age of the solar system is 4.5
billion.years.

(d). Dicke and Peebles (1960)

Early universe had been at least as hot as the Sun center, where He is currently
produced. The hot early universe must therefore have been filled with many high-energy,
short-wavelength photons, which formed a radiation field with that can be given by
Planck’s blackbody law. The universe has expanded so much since those ancient times that
all those short-wavelength photons have their wavelengths stretched by a tremendous factor.
As a result, they have becomes low-energy, long-wavelength photons.

The temperature of this cosmic radiation field is now quite low, only a few degrees above 0
K.

19
(e). Arno Penzias and Robert Wilson

No matter where in the sky they pointed their antenna, they detected faint background
noise.
They had discovered the cooled-down cosmic background radiation left over from the
hot Big Bang.

(f). Cosmic Background

Left over from the hot Big Bang.

T = 2.726 K, cosmic microwave background

0.0029
max  [m] (Wien’s displacement law)
T[K ]

When T = 2.726 K, max  1.06[mm] .

The spectrum of the cosmic microwave background is given by COBE (Cosmic

Background Explorer).

20