Chapter4 PDF
Chapter4 PDF
The speed of a wave propagating through a medium is not a constant. This is especially
true for the non-homogeneous medium, the ocean. The speed of sound through water has been
found to be mainly a function of three factors. They are temperature, pressure or depth and
salinity. Because the speed is not constant, sound does not travel along straight paths.
Temperature
In general, for most areas of the ocean, the water temperature decreases from the surface
to the bottom, but there are many local variations. Shallow layers see the most variation with
time and depth (ie. Surface mixing, solar heating, currents, seasonal variations, etc). In vary
deep water, the temperature eventual becomes constant with depth at about 4 C.
Depth
Hydrostatic pressure makes sound velocity increase with depth because of variations in
the bulk modulus, B. This effect is linear in the first approximation with an increase of 0.017
m/s per meter increase in depth.
Recall in Physics I we showed that pressure varies with depth according to the simple
formula,
P = Po + ρgh
(From: Lurton, X. An Introduction to Underwater Acoustics, 1st ed. London, Praxis Publishing
LTD, 2002, p37)
Salinity
The change in the mix of pure water and dissolved salts effects sound velocity. Salinity
is expressed in practical salinity units (p.s.u.). These unit have the same magnitude as the
traditional parts per thousand (‰). Most oceans have a salinity of 35 p.s.u., although salinity can
vary locally based on hydrological conditions. Closed seas have a greater difference in their
salinity (38 p.s.u. for Mediterranean Sea due to evaporation, 14 p.s.u. for Baltic Sea due to large
freshwater input). Salinity varies very little with depth, but there can be stronger variations near
river estruaries, melting ice, etc.
4-1
Velocity Models
In the 1940’s, sound velocity variations and their affect on acoustic propagation were first
noticed and studied. It is very difficult to locally measure sound velocity, but easy to measure
the parameters that affect it (temperature, salinity, and depth). Several models have been created
to predict sound velocity. A good first approximation is that developed by Medwin (1975). It is
simple but limited to 1000 meters in depth:
1580
1560
1540 ppt salinity
Sound Speed (m/s)
1520 0
1500 30
1480 35
1460 40
1440
1420
1400
1380
0 5 10 15 20 25 30 35 40
Temperature (C)
(From: Lurton, X. An Introduction to Underwater Acoustics, 1st ed. London, Praxis Publishing LTD, 2002, p37)
More recent and accurate models have been developed and include Chen and Millero (1977).
Their model is endorsed by UNESCO and used as the standardized reference model:
4-2
3
c = c0 + c1P+ c2 P + c3P + AS+ BS + CS2
2 3 2
Where,
t - temperature (° C)
z - depth (m)
S - salinity (p.s.u.)
As you can see, the speed of propagation has a very complicated dependence on these
three factors. Some thumbrules that you can use to relate the dependence of the speed of sound
in seawater to each of the factors are:
(From: Principles of Naval Weapons Systems, Edited by Joseph B. Hall, CDR, USN, Dubuque, IA: Kendall/Hunt
Publishing Co, 2000, p.179)
Seawater contains many inhomogenieties, including bubble layers close to the surface,
mineral particles in suspension, and living organisms. These are all potential scatterers of
acoustic waves, especially at higher frequencies.
To predict the direction of propagation of a sound wave in the ocean, we must know the
speed of sound as a function of position (or depth) in the ocean water. To measure the speed of
sound in water, the Navy has developed several tools to measure the temperature of the seawater
as a function of depth or the velocity of sound directly.
4-3
The most widely used tool is an Expendable BathyThermograph or XBT (picture
compliments of ES419). XBTs are launched from submarines, surface ships and even aircraft.
These measure the temperature of the water as the device sinks at a known rate and transmits this
back to the launching platform. This provides a detailed plot of temperature as a function of
depth. Neglecting salinity, the Sound Velocity Profile or SVP can be calculated as a function of
depth and temperature (since these cause the greatest variation in the speed of sound in
seawater.)
Expendable Bathythermograph
Canister Loading Breech
Terminal
Board Depth/Temperature
Chart
RECORDER
Wire Spool
Thermistor
PROBE (XBT)
Many modern submarines are often equipped with velocimeters that calculate the speed
of sound in situ. Other submarines have systems that calculate and record sound speed using
temperature and depth measurements from onboard ships instruments.
Sonic Layer
Depth (LD)
T P
C
Deep Sound
Channel Axis
4-4
Using a Sound Velocity Profile and Snell's Law
We will now shift from thinking of sound as a wave and using the wave equations to
sound as a ray and using Snell’s Law. We can look at either the grazing angles, referenced to the
horizontal and used when looking at refraction, or incidence angles, referenced to the vertical
and used for refraction and backscattering.
In the below sketch, a plane wave is moving towards a boundary beyond which the speed
of sound is much slower. As the wavefronts hit the boundary they slow down and bend more
normal to the boundary. Specific examination of the wave after the right edge hits the boundary
at point A shows that the left side of the wavefront must travel a distance from B to D expressed
as the product of the sound speed c1 and some time interval ∆t. In that same time interval the
right edge of the wave front moves from A to E expressed as the product of sound speed c2 and
Refraction
BD c1∆t
High c1 cos (θ1 ) = =
AD AD
B AE c2 ∆t
BD = c1∆t cos (θ 2 ) = =
AD AD
θ1 θ1
A θ2 D
AE = c2 ∆t cos (θ1 ) cos (θ 2 ) 1
= =
E c1∆t c2 ∆t AD
Low c2
cos (θ1 ) cos (θ 2 )
=
c1 c2
some time interval ∆t. Using trigonometry we see that the ratio of the cosine of the grazing
angle to the speed of sound remains constant across the boundary. This observation is called
Snell’s Law.
Snell's law and ray theory are well suited for each other. Imagine that a sound ray is
transmitted through a series of mediums label 1 through 4 with sequentially increasing sound
speed. In each medium, the angle the ray makes with the horizontal, θ, will depend on the angle
it has in the previous medium and the speed of sound for each medium. The figure below
depicts the relation.
4-5
1 θ1
2 θ2
θ3
3
θ4
4
Notice that when a ray is in a layer and horizontal, θ = 0° and the cos(θ) = 1. We call the speed
of sound when the ray is horizontal, co.
Using Snell's Law from above, we can approximate the behavior of a sound ray as it travels
through a medium where the speed of sound is changing at a constant rate. Let's take the
example where the speed of sound increases as a function of depth as shown.
If the speed of sound increased in each layer as shown, a sound ray would travel in a path
the same as the one already shown.
c1 c2 c3 c4
d 1
e 2
p 3
t 4
h
1
2
3
4
(Notice that the sound ray is bending back towards the layers where the sound
speed is lower. This can be used later to qualitatively determine the ray path for
sound in water.)
More realistically though, the speed of sound changes as a continuous function. If we use a
continuous function instead of the step function for the speed of sound vs. depth, the speed of
4-6
sound as a function of depth can be described by a simple linear equation. This result can be
used to find functions for the radius of the path of the sound ray as well as other quantities.
speed of sound
d 1
e 2
p 3
t 4
h
(Korman, M.S. Principles of Underwater Sound and Sonar, the preliminary edition. Dubuque, IA: Kendall/Hunt
Publishing Company, 1995, pgs 145-147)
The speed of sound, shown as the dotted line, can be expressed as (c1 is the surface temperature):
c = c1 + gz
∆c
where g is the gradient, g = . From Snell's Law and inserting our relation ship for c, yields:
∆z
cos θ1 cos θ
=
c1 c
cos θ1 cos θ2
=
c1 c1 + gz
z = R ( cos θ − cos θ1 )
c1
where R is defined as: R ≡ . Soon we will show R is the radius of curvature of the
g cos θ1
sound ray. θ is always measured clockwise from the horizontal axis.
x1 x2
R
z1 c1
θ1
z2 c2
θ2
Positive gradient, g
4-7
In polar coordinates we know that the slope of a line is
dz
= tan θ
dx
From above we see that dz = −R sin θdθ . To find how the ray angle varies with distance x,
sin θ
dx = − R dθ = − R cos θdθ
tan θ
Integrating both sides gives the result that:
x − x1 = −R [sin θ − sin θ1 ]
Integrating both sides of dz gives:
z − z1 = R [ cos θ − cos θ1 ]
Rearranging these two equations:
x − x1 − R sin θ1 = −R sin θ
z − z1 + R cos θ1 = R cos θ
Or
x − x p = −R sin θ
z − z p = R cos θ
With
x p = x1 − R sin θ1
z p = z1 + R cos θ1
Squaring the top two equations and adding the results gives the equation of a circle,
(x − x ) + (z − z )
2 2
p p = R2
c1
Specificically, the circle has radius, R ≡ , and is centered at the point (xp,zp). Thus we
g cos θ1
have shown that a sound ray in a layer of constant sound speed will travel along the arc of a
circle.
4-8
To sum up the results then:
∆c Gradient
g=
∆z
(How fast the speed of
sound changes per meter
change in depth.)
cn Snell's Law
= constant
cos θ n
These equations will only work for one specific sound ray emanating from a source in an
environment with a constant gradient. The last two equations in the table are presented without
proof, but are useful results from many standard sources.
Example 1
4-9
csurface=1500 m/s
θI=20°
θII=30°
Sound leaves the ship at two different angles, θ1 and θ2. Note the path travelled by each
ray is different and if we calculate the parameters R, ∆x and ∆z, each of these will be different
for each ray.
and
c 1500 m/s
R II = =
g cos θII ( 0.1 sec-1 ) ( cos 30° )
R II = 17,300 meters
The skip distance, X, is the distance between successive places where the sound ray
stikes the surface. The easiest way to calculate this is to calculate the displacement, ∆x, from
where the sound strikes the surface first to where the sound has leveled off or gone horizontal (θ2
= 0°). Thus:
X = 2∆x = −2 R(sin 0° − sin θ )
X = 2 R sin θ
where θ is the angle of reflection from the surface. So for each ray:
4-10
X I = 2(16,000 m )sin 20°
X I = 11,000 m
and
X II = 2(17,300 m )sin 30°
X II = 17,300 m
The results of the calculations for each ray are significantly different from each other and
show how the ray paths depend on the initial angle of the ray.
We can do the same for the depth the rays to. The maximum depth excursion of the ray
below its starting depth occurs when the ray goes horizontal again (θ2 = 0°) or:
∆z max = R(cos 0° − cosθ )
∆z max = R(1 − cosθ )
Example 2
c1 θ1
θ2 0°
c2
c0
Use the figure above and the following information to answer the questions.
1
From: Korman, M. S. Principles of Underwater Sound and Sonar, the preliminary edition, p. 144.
4-11
b. Determine c0.
Again using Snell's law and the θ0 = 0°
c2
= c0
cos θ 2
1299 m / s
c0 = = 1500 m/s
cos30°
4-12
Problems
of the ocean?
c) If a ray reflects off the
surface of the ocean at an
angle of 2.15° (assume
the surface is perfectly
flat), what will be the
100
skip distance of the
sound ray?
d) This is an example of:
i) a positive gradient
ii) a negative gradient
e) If a sub is at 50.0 m, what is the largest angle below the horizontal where the ray will not
reach 100. m?
f) What is the skip distance of the limiting ray?
4-13
Use the following SVP for the remaining problems:
500
2000
Depth (m)
2500
3000
3500
4000
4500
c3
5000 Bottom
3. Compute the sound speed gradients for 0 < z < 1200 m and 1200 < z < 4800.
4. A ray starts at 1525 m with a grazing angle if 15 degrees (recall a positive grazing angle is
pointed below the horizontal).
a) What are the sound speeds at depths of 1525 m and 2440 m?
b) Does the ray curve upward or downward?
c) What is the grazing angle at 2440 m?
d) At what sound speed will the ray become horizontal (a horizontal ray has a grazing angle
of 0 degrees)?
4-14
5. A sound source is at a depth of 1200 m.
a) At what angle with respect to the horizontal does the ray have to make at 1200 m so that
that when it reaches the surface the grazing angle is 0 degrees? This is called the surface
limiting ray.
b) What angle with respect to the horizontal does a ray have to make at 1200 m so that when
it reaches the bottom at 4900 m, the angle is 0 degrees? This is called the bottom limiting
ray.
c) At what depth below 1200 m is the sound speed equal to that at the surface?
d) At what angle with respect to the horizontal does a ray have to make at 1200 m so that
when it reaches the the depth found in c), the grazing angle is 0 degrees? This is called
the lower limiting ray
e) Compute the radius of the surface limiting ray.
f) Compute the radius of the bottom limiting ray.
g) Compute the radius of the lower limiting ray.
h) Compute the horizontal distance that the bottom limiting ray travels from the source until
it grazes the bottom.
i) Compute the horizontal distance that the surface limiting ray travels from the source until
it just grazes the surface.
6. A ray leaving a sound source at 1200 m points downward with an angle of 30 degrees with
respect to the surface.
a) How far will it travel horizontally until its angle with the horizontal is 25 degrees?
b) At what depth does the ray in a) make an angle of 25 degrees with respect to the
horizontal.
4-15
Lesson 4
Depth
Depth
Depth
Depth
Depth
0 ≤ z ≤ 1000 meters
Sound Speed (m/s)
1580 1480
1560
Series1
Sound Speed (m/s)
1
Lesson 4
=
where c1 < c2 < c3 < c4 and θ1 > θ2 > θ3 > θ4 c1 c z
cos θ1 cos θ
cos(θ1 ) cos(θ 2 ) cos(θ3 )
= = = .... =
cos(θ n )
= constant =
c1 c2 c3 cn
c1 c1 + gz
c1
z = R ( cos θ − cos θ1 ) R=
g cos θ1
x1 x2 csurface=1500 m/s
c1
θI=20°
z R=
g cos θ1
x
z − z1 = R ( cos θ − cos θ1 )
Positive gradient, g
2
Lesson 4
( x − x ) + ( z − z ) = ( −R )
2 2
sin 2 θ + R 2 cos 2 θ
2
p p
x − x1 = − R ( sin θ − sin θ1 ) (x − x ) + (z − z ) p
2
p
2
= R2
θ1
z − z1 = R ( cos θ − cos θ1 ) z − z1 = R ( cos θ − cos θ1 )
z1
R c1
z1 c1 θ1 R
z2 c2 c1 c1
θ2 R= R=
g cos θ1 g cos θ1
Positive gradient, g ∆c c − c1 ∆c c − c1
g= = Negative gradient, g g= =
∆z z − z1 ∆z z − z1
Example 1 Example 2
csurface=1500 m/s
θI=20°
θ1
c1 θII=30°
θ2 0°
c2
c0
c100 m=1510 m/s
3
Lesson 4
c1
z = R ( cos θ − cos θ1 ) R=
g cos θ1
Slope = tanθ
x1 x2
z1
θ
z2
z 2 − z1 ∆z dz
tan θ = = =
x 2 − x1 ∆x dx