Lecture 10: Tidal Power

Chris Garrett

1 Introduction
The maintenance and extension of our current standard of living will require the utilization
of new energy sources. The current demand for oil cannot be sustained forever, and as
scientists we should always try to keep such needs in mind. Oceanographers may be able
to help meet society’s demand for natural resources in some way.
Some suggestions include the oceans in a supportive manner. It may be possible, for
example, to use tidal currents to cool nuclear plants, and a detailed knowledge of deep
ocean flow structure could allow for the safe dispersion of nuclear waste. But we could also
look to the ocean as a renewable energy resource. A significant amount of oceanic energy is
transported to the coasts by surface waves, but about 100 km of coastline would need to be
developed to produce 1000 MW, the average output of a large coal-fired or nuclear power
plant. Strong offshore winds could also be used, and wind turbines have had some limited
success in this area.
Another option is to take advantage of the tides. Winds and solar radiation provide the
dominant energy inputs to the ocean, but the tides also provide a moderately strong and
coherent forcing that we may be able to effectively exploit in some way. In this section, we
first consider some of the ways to extract potential energy from the tides, using barrages
across estuaries or tidal locks in shoreline basins. We then provide a more detailed analysis
of tidal fences, where turbines are placed in a channel with strong tidal currents, and we
consider whether such a system could be a reasonable power source.

2 Barrages and Tidal Locks

117
 Mean Tidal Amplitude (m) Basin Area (km2 )
La Rance, France 4 17
Bay of Fundy, Canada 5.5 240
Annapolis Royal, Nova Scotia 3.2 6
Severn Estuary, Great Britain 4 420
Garolim Bay, South Korea 2.5 85

The only major barrage plant is located at La Rance, Brittany, on the northern coast of
France, which began operation in 1967. It has a capacity of 240 MW making it a significant
energy source, given that typical coal and nuclear plants have an output of about 1000 MW.
A disadvantage of this method is that the power source is clearly an intermittent one, which
is why there is a compromise between releasing the energy at the lowest tide and at the
period of greatest demand. La Rance was supposed to be the first of many tidal plants
across France but their nuclear program was greatly expanded around this time, and so
La Rance remains the only major tidal power plant in the world today. Although it has had
an impact on the local ecology, it has been argued that the ecology was not particularly
unique to begin with.
The only other commercially viable barrage plant is the Annapolis Royal station in Nova
Scotia, which was activated in 1982 and has a capacity of about 20 MW. It was designed as
an experimental station that would lead to an array of tidal power stations across the head
of the Bay of Fundy, which is not only very large, but it also possesses the largest tides in
the world. Such a network may be able produce an energy output as large as 5000 MW,
but the environmental consequences could be severe; the more dire of predictions suggest
that it could lead to a 5% drop in the Bay of Fundy tidal levels and a 10% rise in New
England’s levels. In any case, a better understanding of the effects of such a network would
be needed before such an ambitious project can be attempted.
Other barrage plants exist, such as in Russia and China, but none have an output greater
than about 500 kW. But the most dramatic of all proposed tidal plants is for a barrage
across the Severn Estuary, between Wales and England. There is the possibility that over
8000 MW of power could be extracted, which could supply 12% of the current UK energy
demand. Evaluations were conducted from 1974 to 1987, but the project was eventually
shelved because of economic and environmental concerns. Other schemes, similar in size to
the one of La Rance, may be implemented soon in South Korea.

3 Tidal Fences
Instead of directly using the variable potential energy of the oceans, it may be possible
to utilize the strong tidal currents that are generated through narrow channels in certain
parts of the world. We will attempt to quantify the actual amount of power that could
be extracted from such a plant. This analysis has been done previously by Garrett and
Cummins.[1]
We consider flow through a channel of variable cross-section (figure 1). The current
speed u(x, t) is assumed to be function of time t as well as x, but independent of the
cross-channel position. The dynamical equation governing the flow is

118
 Ae
ue

A(x) x
u

Figure 1: A channel connects two basins with different tidal elevations. At any instant, the current
through the channel has speed u, a function of the local cross-sectional area A(x), where x is the along-strait
coordinate. At the channel entrance, the current enters from all directions, but at the exit it may separate
as a jet with speed ue , surrounded by comparatively stagnant water with elevation the same as that in the
downstream basin.

∂u ∂u ∂ζ
+u +g = −F (1)
∂t ∂x ∂x
where the slope of the surface elevation ζ provides the pressure gradient to drive the flow and
F represents an opposing force associated with natural friction and possibly the presence
of turbines (in a uniform “fence” across the flow, a more efficient scheme than isolated
turbines [2]).
If the channel is short compared with the wavelength of the tide (generally hundreds
of kilometers even in shallow water), volume conservation implies that the flux along the
channel Au = Q(t), independent of x. (We neglect small changes in A associated with the
rise and fall of the tide.) Using this in (1) and integrating along the channel implies
Z L
dQ 1
c − gζ0 = − F dx − ue |ue | (2)
dt 0 2
RL
where c = 0 A−1 dx and ζ0 (t) is the sea level difference between the two basins, assuming
that this difference is unaffected by the flow through the channel. We note that, if A
increases rapidly at the ends of the channel, then the geometrical factor c is insensitive
to their locations at x = 0, L. We allow for flow separation at the channel exit, with u e
denoting the current speed there.
We start by assuming that the natural frictional term and the head loss associated with
separation at the exit are small, so that the natural regime has a balance between the sea
level difference and acceleration. For ζ0 = a cos ωt, a sinusoidal tide with amplitude a and
frequency ω, Q = Q0 sin ωt where Q0 = ga(ωc)−1 . The power generated at the turbines is
the integral along the channel of the product of water density ρ, current u, cross-section A,
and the local frictional force F representing the turbines. Thus the average power extracted
from the flow by the turbines is
Z L Z L
P = ρF Q dx = ρQ F dx, (3)
0 0
with the over-bar indicating the average over a tidal cycle. This would need to be
multiplied by a turbine efficiency factor to give the electrical power produced. We first

119
assume that the dragR associated with the turbines is linearly proportional to the current.
L
Then we may write 0 F dx = λQ and P = λρQ2 , where λ is related to the number of
turbines and their location along the channel. The governing equation
dQ
c − ga cos ωt = −λQ (4)
dt
is easily solved to give
1 2 2
2 ρλg a
  
ga −iωt
Q = Re e , P = . (5)
λ − icω λ2 + c 2 ω 2
As expected, P at first increases with λ (more power is generated as more turbines are
added), but then decreases as too many turbines choke the flow. The maximum power,
obtained when λ = cω, is
1 1
Pmax = ρg 2 a2 (cω)−1 = ρcωQ20 . (6)
4 4
The flow through the channel is then reduced to 2−1/2 , or 71%, of its original value. We
also note that this maximum power is independent of the location of the turbines along the
channel (with the fewest being needed if they are deployed at the smallest cross-section).
This result is independent of the representation of the turbine drag.
A more realistic representation of the turbine drag would be quadratic in the current
speed. We have considered an arbitrary exponent, by non-dimensionalizing with t = ω −1 t0 ,
Q = Q0 Q0 , numerically solving
dQ0
− cos t0 = −λ0 |Q0 |n−1 Q0 (7)
dt0
and evaluating P/Pmax = 4λ0 |Q0 |n−1 Q02 as a function of λ0 for different values of n (figure
2). The maximum power for n = 2 (quadratic drag) is very close to the value P max derived
for n = 1, and it does not seem that any other drag law could lead to more power.
We thus take Pmax as a reasonable estimate of the maximum power available in the case
of negligible background friction and exit separation effect. We may evaluate it in terms of
conditions at the constriction with smallest cross-sectional
R L area, A min , where the amplitude
−1
of the undisturbed tidal current is umax , if we write c = 0 A dx = Leff /Amin , noting that
this effective length Leff is likely to be considerably less than the total channel length L.
Then
1
Pmax = ρωLeff Amin u2max . (8)
4
This may be compared with the reference value
1 2
P0 = ρAmin |umax cos ωt|3 = ρAmin u3max (9)
2 3π
which gives the average undisturbed kinetic energy flux through the channel and is usually
used as an estimate. The ratio is
Pmax 3π ωLeff
= (10)
P0 8 umax

120
Figure 2: The scaled maximum power as a function of a parameter λ0 representing the number of turbines
in the channel, for various values of n, where the turbine drag is assumed proportional to the nth power of
the current speed.

and can either be less than 1 (short channel, strong currents) or greater than 1 (vice
versa). For example, with ω = 1.4 × 10−4 s−1 , as for the semi-diurnal tide, Leff = 5 km,
and umax = 3 m s−1 , then Pmax /P0 = 0.3, so P0 would overestimate the resource. With
these values and Amin = 5 × 105 m2 (as for a width of 5 km and a depth of 100 m), then
Pmax = 800 MW.
The calculations so far have neglected background friction and separation effects. If
these can be lumped together in a single term, as for quadratic friction in the channel, and
represented by a value λ0 in (7), we may still solve (7) but the power from the turbines is
now (λ0 − λ0 )/λ0 times the value calculated with λ0 = 0. Here we examine the situations in
which the acceleration is much less important than friction in the channel (as for a shallow
channel) or the separation effect (as for a short channel). The momentum equation is then
Z L
−gζ0 = − Fturb dx − αQ|Q|, (11)
0
RL
giving a balance at any instant between the surface slope, the drag associated 0 Fturb dx
with the turbines, and the internal friction and separation, where
Z L
1
α= Cd (hA2 )−1 dx + Ae−2 . (12)
0 2
Here Cd is the drag coefficient, h the water depth, and Ae the cross-sectional area at the
exit.
In the absence of turbines, the volume flux is |Q0 | = (g|ζ0 |/α)1/2 . With turbines, the
instantaneous power is ρQ(gζ0 − αQ|Q|). This has a maximum of
Pmax = (2/33/2 )ρQ0 gζ0 = 0.38 × ρQ0 gζ0 (13)
where we have also averaged over a cycle. The flow at any instant is 3 −1/2 = 0.58 of what
it was in the absence of the turbines, and 2/3 of the head loss along the channel is now
associated with their operation.

121
 If the friction within the channel is much less important than the separation effect, (13)
may be written
1
Pmax = 0.38 × ρAe |ue0 |3 (14)
2
where ue0 = (2gζ0 )1/2 is the exit speed at any instant in the natural regime. This formula
is similar to (9), but we note the reduction factor of 0.38 and, particularly, the fact that
it must be evaluated at the exit, likely giving a considerably lower value than would be
obtained at the most constricted part of the channel. The turbines could still be deployed
there, however.
In the preceding analysis we have assumed that the current is independent of the cross-
sectional position. If there is shear vertically and across the channel, we may deal with the
average speed in the acceleration term, but there will be minor changes in other parts of the
theory. We have also ignored the back effect of changes in the channel flow on the forcing
tides. This is likely to be small if the basins are large and deep, but there may be a positive
feedback which will increase the head as turbines are introduced. This will slightly increase
the available power.
Further consideration for particular sites will require more detailed analysis, but (8),
(13) or (14), whichever is appropriate, will still give a good preliminary estimate of the
power potential of a number of suggested sites around the world. We emphasize that the
commonly used estimate, based on the energy flux in the natural state through the smallest
cross-section, is incorrect and likely to be an overestimate. In general, significant power
potential depends not only on the obvious factors of strong currents and large cross-section,
but also on either the channel being long, or the currents at the exit separation, rather than
just the constriction, being large.

Notes by Viswewaran Nageswaran and Marshall Ward.

References
[1] C. Garrett and P. Cummins, “The power potential of tidal currents in channels,” Proc.
Roy. Soc. A (2004), submitted.

[2] C. Garrett and P. Cummins, “Generating power from tidal currents,” Journal of Wa-
terway, Port, Coastal and Ocean Engineering 130, 114 (2004).

122