Water Power and Dam Construction. Issue October 2010
Подождите немного. Документ загружается.
WWW.WATERPOWERMAGAZINE.COM OCTOBER 2010 11
NEW TECHNOLOGY
PRODUCTS AND APPLICATIONS
Power Jacks goes with the flow
Power Jacks products are helping to control the ow of a river – and
support a major hydroelectric scheme in the process – in one of the
most scenic areas of Scotland.
The company has provided a 100kN S-Series rotating screw jack
in a ram style linear actuator design together with motor and reduc-
tion gearbox, for a weir system on the River Morar, near Mallaig on
Scotland’s west coast.
The river, only a few hundred metres in length, forms the link
between the extensive waters of nearby Loch Morar – which reaches
depths of over 300m – and the sea.
The weir, constructed a short distance from a hydroelectric dam
and sh ladder at the west end of Loch Morar, is a pivotal element of
the wider water ow systems at the locality.
The screw jack essentially performs the function of connecting with the
weir dam gate, raising or lowering the gate depending on the river level.
The unit has worked successfully since its installation in 2001. Since
then it has had one upgrade where the electric motor was replaced
with an alternative unit with integral hand wheel.
The screw jack operates over a stroke of 2350mm, pivoting on its
integral trunnions located midway on the ram's outer tube. At the end
of the ram a clevis end with spherical bearing connects the unit to the
weir gate. Originally the screw jack was driven by a 2.2kW geared
motor with position feedback coming from end of stroke limit switch-
es and a absolute encoder that links back to the control system.
http://www.powerjacks.com
Rock Fall (UK) Ltd has released its Tomcat range of Total Protection
Footwear, the rst to incorporate internal metatarsal protection to
meet European S3 Safety Standards, the company says.
Based in Alfreton, Derbyshire, the designer and manufacturer
of protective safety footwear has launched this new range to help
reduce the number and severity of metatarsal injuries, a widespread
and extremely painful, debilitating injury to suffer. Designed for the
comfort and safety of personnel working in industrial environments,
the boots were developed through extensive research and testing to
replace the traditionally heavy, uncomfortable, unsightly and expen-
sive boots currently on the market, says the company.
The range has taken more than two years to develop and surpasses
the EN ISO 20345:2004/A1 2007 European Safety Standards. The
main design feature is the patented, integral metatarsal protector: a
lightweight, exible inner plate that protects the wearer’s fragile meta-
tarsal bones from impact. Throughout the development of the boots,
Rock Fall also registered ve design features for protection. The foot-
wear is manufactured to the highest S3 specication and also protects
the user’s toes and sole with a metal toecap and steel mid-sole.
Using full grain leather and Rock Fall’s unique ‘Activ-step’
foot-bed for comfort, the boots also feature rubber outsoles
constructed for the
highest durabil-
ity and are able to
withstand heat up to
300 ˚C. Developed
with the wearer and
their specic work-
ing en viron ment
in mind, the range
also includes a fully
waterproof design.
http://rockfall.co.uk
Dutch refrigeration specialist, N.R. Koeling, and US-based industrial
ice equipment manufacturer, North Star Ice Equipment Corporation,
have won a contract to provide industrial concrete cooling equipment
for the expansion of the Panama Canal.
The contract was awarded by GUPC, the consortium in charge of
designing and building a new set of locks as part of the Panama Canal
Expansion Project. The project will create a new lane of trafc and
double the canal's capacity.
N.R. Koeling will supply two large ice plants to provide ake ice for
the concrete batching and mixing process, one on the Atlantic Ocean
side of the Panama Canal and one on the Pacic side. This will include
a wet belt cooling system, an ice making and delivery plant, a sand
cooling plant, a chilled water plant and an air blast cooling system.
The industrial ice making, handling and storage equipment in the
plant is supplied by North Star Ice Equipment. This includes 12 of
North Star’s largest ake ice makers, four North Star modular ice
rakes, and eight day tanks, which will be used to meter the ice for
each batch of concrete.
The Panama Canal Expansion involves the construction of a third
set of canal locks, which will enable the Canal to handle twice the com-
mercial trafc it handles today. In addition, the expansion will allow
the Canal to accommodate larger “Post Panamax” container vessels,
which are now being built. The project includes the construction of two
new lock facilities– one on the Atlantic side and another on the Pacic
side – each with three chambers and three water reutilization basins; the
excavation of new access channels to the new locks and the widening
of existing navigational channels; and, the deepening of the navigation
channels and the elevation of Gatun Lake’s maximum operating level.
http://www.nrkoeling.nl
http://www.northstarice.com
Best foot forward
Cooling equipment
12 OCTOBER 2010 INTERNATIONAL WATER POWER & DAM CONSTRUCTION
INSIGHT
I
T is fairly obvious that water power
depends on the availability of water.
That likely availability is vital for sizing
installed turbine capacity. It is equally
true to say that the sizing of spillways depends
on reliably estimating ood size. Long term
yield and likely oods are both determined on
the basis of historic records and conventional
wisdom analyses these assuming they are
stochastic, or random. Yields are assessed on
the basis of mean long-term values and, say
95% reliability.
In practice wouldn’t it be good if we could
predict not just long-term averages but also
when the wet years were going to arrive and
also the dry years? Well, perhaps we can.
Some years ago the writer analysed the
levels of Lake Victoria in east Africa as part of
a major refurbishment project. This revealed a
clear cyclic behaviour in lake level and hence in
discharges from the lake down the Nile system
and up into Egypt. More recently the writer
has been working on the potential yield of a
proposed major hydropower development in
Zambia. Again clear cyclic patterns were evi-
dent in the record. In the course of working on
both these projects it became clear that others
had also identied cyclic patterns elsewhere in
Africa and in other parts of the world.
In the short to medium term horizons
required for financial planning purposes
on hydropower projects, a recent study by
Rydgren et Al (2007) for the World Bank
concluded that natural variability would
dominate over any longer term trends due to
man-made climate change. This makes the
identication of such natural patterns espe-
cially useful and valuable.
A recent paper included in the Proceedings
of the Institution of Civil Engineers (Mason,
2010), focused on African correlations. This
article is based on the ICE paper but also
gives some other examples.
Lake Victoria – East Africa
At 67,000km2 and bordered by Kenya,
Uganda and Tanzania, Lake Victoria is the
third largest lake in the world. Its catchment
is nearly the size of the UK. From 1900 to
around 1960, Lake Victoria levels oscillated
fairly constantly about a Jinja gauge level
of +11m. But a dramatic rise in lake levels
throughout East Africa occurred in the early
1960s, including those of Lake Victoria.
The lake discharges into the Nile system
via the Owen Falls hydropower station. The
writer was involved in refurbishing the works
at Owen Falls in the 1990s and revisited an
apparent solar correlation with Lake levels
observed back in the early 1900s. The result-
ing analyses showed not only a falling lake
level after 1964 but also an oscillation around
that falling trend which, again, appeared to
demonstrate a solar correlation, (Mason
1993, 2006a) [see Figures 1 and 2].
The falling trend in Lake level meant that
the new Kiira hydropower station commis-
sioned in 2000 did not, in fact, produce any
long-term, additional energy for Uganda.
The predictable fall in Lake levels coupled
with some over-abstraction led to a tempo-
rary halt of generation at Kiira in 2007.
If the historic cyclic patterns of Lake behav-
iour continue, a lake level oscillation around
mean annual Jinga gauge levels of approx
+10.65m and +11.25m will correspond to 11
year cyclic outflows varying between about
500 and 730m
3
/sec. For a scheme with an
operating head of about 22m it has been dem-
onstrated (Mason 2006b) that these will con-
vert into signicant annual revenue differences.
Any new hydropower generation coming on
stream during an upward trend in lake level
will pay back investment costs approximately
two years earlier than if generation commenced
on a falling trend, (see Figure 3).
Kafue River – Zambia
A recent study by the writer demonstrated
that 20-year mean ows in the Kafue River
in Zambia corresponded well to reconstruct-
ed rainfall records based on regional tree ring
How predictable are water resources?
Dr Peter Mason, technical director of international dams and hydropower at MWH, reveals
how some water resources might be more predictable than generally supposed
13.00
12.00
11.00
10.00
Jinja gauge levels (m)
12.40
11.40
10.40
180.0
150.0
120.0
90.0
60.0
30.0
0.0
Lake (Jinja gauge) level (m)
Solar variations (sunspot number)
Approx mean level
prior to 1962
Falling trend of
29mm per year
Lake rise from
1962 to 1964
1900 1910 1920 1930 1940 1950
Year
1960 1970 1980 1990 2000
0
0
0
11
11
11
999
666
333
0
0
0
1900 1910 1920 1930 1940 1950
Year
1960 1970 1980 1990 2000
Period
1896-1928
Period
1968-2005
Natural
lake level
Sunspot
number
Lake level minus
the 29mm per
year falling trend
Simulated level sequence
11.30
11.20
11.10
11.00
10.90
10.80
10.70
10.60
0 5 10 15 20 25 30 35
Commission on
rising trend:-
10.2 year payback
at 15% rate
Commission on
rising trend:-
12.1 year payback
at 15% rate
Year
13.00
12.00
11.00
10.00
Jinja gauge levels (m)
12.40
11.40
10.40
180.0
150.0
120.0
90.0
60.0
30.0
0.0
Lake (Jinja gauge) level (m)
Solar variations (sunspot number)
Approx mean level
prior to 1962
Falling trend of
29mm per year
Lake rise from
1962 to 1964
1900 1910 1920 1930 1940 1950
Year
1960 1970 1980 1990 2000
0
0
0
1
1
1
1
1
1
9
9
9
6
6
6
3
3
3
0
0
0
1900 1910 1920 1930 1940 1950
Year
1960 1970 1980 1990 2000
Period
1896-1928
Period
1968-2005
Natural
lake level
Sunspot
number
Lake level minus
the 29mm per
year falling trend
Simulated level sequence
11.30
11.20
11.10
11.00
10.90
10.80
10.70
10.60
0 5 10 15 20 25 30 35
Commission on
rising trend:-
10.2 year payback
at 15% rate
Commission on
rising trend:-
12.1 year payback
at 15% rate
Year
Above: Figure 1 - Levels of Lake Victoria from 1896 to 2005; Figure 2 – Levels of Lake Victoria from 1896 to
1928 and from 1968 to 2005 compared to solar variation in the form of sunspot number indices, but with
the 29mm per year falling trend in lake level eliminated from the 1968 to 2005 data
WWW.WATERPOWERMAGAZINE.COM OCTOBER 2010 13
INSIGHT
records. The tree ring records extended back
200 years and this allowed the identication
of clear cyclic behaviour. The work was car-
ried out as part of an IFC funded study (Kafue
Gorge Lower Hydroelectric Project, 2009).
The results of the Kafue correlations are
shown in Figure 4. They show a dominant
period of 70 years accounting for about 50% of
the correlation followed by a secondary period at
130 years accounting for about 25%. Tertiary
periods at 35 and 44 years account equally for
the remainder.
The relationships shown in Figure 4 allow
hypothetical mean annual rainfalls to be
extended into the future. This suggests that
these can be used to estimate future long-term
trends in rainfall, river ow and hence hydro
generation capability. Indications are for
ows to increase over the next few decades,
meaning the increasing trends in river ow
may occur sooner than implied by Figure 4.
Africa generally
Frossard E. et al (2006) explored the mean
annual flows in major rivers throughout
Africa between 1945 and 2000. They also
explored the wider record for the Zambezi
between 1905 and 1995, reflecting the
broad underlying trends shown in the pre-
ceding section. Frossard demonstrated a
generally increasing trend of African river
ows between 1950 and 1970 followed by a
decreasing trend thereafter (Figure 5).
Frossard highlighted the need for under-
Examples of cyclic variation in meteorological and proxy weather records
Significant meteorological periodicities (years)
Southern oscillation 3-3.8 6 10-12
Central England temperature 3.1 14.5 23 76
South African rainfall 3.5 10-12 18 21
South American rainfall 3.8 7 20
Extreme 1 hr rainfall (UK) 7.2 10.8 20 50
US east coast temperature 9 20
North Atlantic pressure 7-9 14 50-90
Atlantic hurricanes 9.3 11 15 22 52
Beijing rainfall 9.9 18.6 56 84 126
US rainfall 11 18.6
East African Lakes 11
Nile Floods 18.4 53 77-88
Global air temperature 22 65-70
Significant proxy data periodicities (years)
Galapagos corals 3 6-8 11 17 22 34
Oman speleothem 3 7.5 24 87 134 205 779
Chinese speleothem 3.3 5-6 10-12 14-18 133 194
Zimbabwe tree rings 3.4 9-10 20 35-44 70 130
Geological sediments 5.6 11 22 90 200
Greenland ice cores 6.3 11 20-22 90 200 1470
Fennoscandian tree rings 11 31-34 86-100
US drought index (tree rings) 18.6 22 90
Finnish tree rings 23 30 90
Irish speleothem 78 169 625
Bristlecone pine tree rings 114 208 667 1400
Ocean sediments 550 1500
Table constructed based on data from; Alexander, W. (2007), Burroughs, W.J. (2003), Mason P.J. (1998), Mason P. J. (2006a), Ruzamaikin A. et al. (2006) and the Kafue Gorge Lower
Hydroelectric Project, Climate Change Impact Draft Report (2009).
13.00
12.00
11.00
10.00
Jinja gauge levels (m)
12.40
11.40
10.40
180.0
150.0
120.0
90.0
60.0
30.0
0.0
Lake (Jinja gauge) level (m)
Solar variations (sunspot number)
Approx mean level
prior to 1962
Falling trend of
29mm per year
Lake rise from
1962 to 1964
1900 1910 1920 1930 1940 1950
Year
1960 1970 1980 1990 2000
0
0
0
11
11
11
999
666
333
0
0
0
1900 1910 1920 1930 1940 1950
Year
1960 1970 1980 1990 2000
Period
1896-1928
Period
1968-2005
Natural
lake level
Sunspot
number
Lake level minus
the 29mm per
year falling trend
Simulated level sequence
11.30
11.20
11.10
11.00
10.90
10.80
10.70
10.60
0 5 10 15 20 25 30 35
Commission on
rising trend:-
10.2 year payback
at 15% rate
Commission on
rising trend:-
12.1 year payback
at 15% rate
Year
Below: Figure 3 – In the case of a simulated cyclical level and flow variation sequence for Lake Victoria,
commissioning a hydropower station on a rising trend can lead to a two year improvement in scheme pay-
back time as compared to commission on a falling trend
14 OCTOBER 2010 INTERNATIONAL WATER POWER & DAM CONSTRUCTION
INSIGHT
standing such long-term trends in hydrology
by sighting the design of the Kariba dam. The
original spillway design capacity for Kariba
was shown to be inadequate by the arrival of
two massive floods during its construction.
These led to a re-evaluation of the hydrology
as the dam was being built and the original
design outow capacity of 6000m3/sec using
four gated openings being changed to six gated
openings giving 9000m3/sec.
Rio Parana – South America
The Rio Parana has a catchment area of
3,100,000km
2
and a mean stream ow of
21,300m
3
/sec. It is the fth largest river in
the world by drainage area and the fourth
largest by stream flow. It collects water
from Brazil, Paraguay, Bolivia, Uruguay
and Argentina. Its eventual outlet into the
Atlantic Ocean is via the Rio de la Plata and
the Salto Grande hydroelectric scheme.
Studies by Mauna and Flamenco (2005)
analysed annual river ows from 1900 to
2000. They subtracted annual ow values
from 11 year running mean values and then
carried out a similar exercise for solar vari-
ation using sunspot numbers. The resulting
values for both ow difference and sunspot
number difference were then normalised
by subtracting the mean values and divid-
ing by their respective standard deviations.
The results are shown on Figure 6 and show
a clear correlation between river ow and
solar variation. They rened the comparison
still further by also demonstrating a correla-
tion with shorter term El Nino effects.
In a later paper (Mauna and Flamenco, 2008)
they repeated the exercise using not only sun-
spot numbers but also reconstructed solar irra-
diance values, with equally convincing results.
While the results can be used for assessing likely
future generation potential at the Salto Grande
station they can also be used to assess years of
likely highest future ood risk.
Ireland
A paper by Alexander et Al (2007) records
studies by Nico Willemse. Willemse studied
the departure of Irish rainfall from the long-
term mean precipitation values. He used data
from a number of meteorological stations
in the Republic of Ireland including Dublin,
Malin Head, Valencia and Birr. The selection
ensured a complete geographic coverage from
north to south and included more than 46,000
data sets, some extending back to 1881.
Willemse compared the cumulative depar-
ture of annual rainfall values with an equiva-
lent departure of sunspot numbers from the
cumulative mean. The results are shown on
Figure 7 for Dublin. They demonstrate a
clear correlation between rainfall variation
and solar variation in the Republic of Ireland
from 1881 to 2005.
England
Some years ago, Brooks and Carruthers
(1953) identified an annual rainfall perio-
dicity in England of 51.7 years. They noted
that individual wet years were still possible at
the dry point in this cycle and individual dry
years were still possible at the wet point the
cycle. But what they demonstrated was that
the probability of a wet year was almost twice
that of a dry year near the 51.7 year maximum
while near the minimum it was less than half.
Further studies on the variations of extreme
hourly rainfalls in the United Kingdom were
made much later by May and Hitch (1989).
They identied clear base cycles in the rain-
fall record of 7.2, 10.8, 20 and 50 years. It
can be assumed that the 50 year cycle cor-
responded broadly to the 51.7 year cycle
identied earlier by Brooks and Carruthers.
However they also noted that the 7.2 and
10.8 year cycles combined to give apparent
cycles of nine and 13 years.
Worldwide
The sections above identify periodicities
in the rainfall and/or river run-off from
Africa, South America and the British Isles.
Similar periodicities are indicated in records
from many other parts of the world such
as China and North America. These have
been explored with great detail by Burroughs
(2003) and are shown in Table 1 from the
paper by Mason (2010).
However, detailed hydrological records
are often not available and so in order to
identify historic climatic variation we can use
proxy data, such as lake sediments and tree
rings. The results from such data are shown
on the second part of Table 1. This sum-
Period
Power
Lag
35.1
1.15
17.89
44.0
1.59
28.33
70.0
6.97
11.92
130.0
3.08
39.62
Model 1
20 year average
RSQ 0.87
600
580
560
540
520
500
480
460
440
420
400
Reconstructed annual rainfall (mm)
1800 1850 1900 1950 2000 2050 2100
Year
Senegal at Bakel
Volta at Akosombo
Blue Nile at Rosieres
Oued Hammam at Bou Hanifa
Niger at Tossaye
Zambezi at Victoria Falls
Chari at N’Djamena
Niger at Niamey
12
10
8
6
4
2
0
-2
Deviation of cummulative averaged annual river
flow volumes from long term mean values
1940 1950 1960 1970
Year
1980 1990 2000
Period
Power
Lag
35.1
1.15
17.89
44.0
1.59
28.33
70.0
6.97
11.92
130.0
3.08
39.62
Model 1
20 year average
RSQ 0.87
600
580
560
540
520
500
480
460
440
420
400
Reconstructed annual rainfall (mm)
1800 1850 1900 1950 2000 2050 2100
Year
Senegal at Bakel
Volta at Akosombo
Blue Nile at Rosieres
Oued Hammam at Bou Hanifa
Niger at Tossaye
Zambezi at Victoria Falls
Chari at N’Djamena
Niger at Niamey
12
10
8
6
4
2
0
-2
Deviation of cummulative averaged annual river
flow volumes from long term mean values
1940 1950 1960 1970
Year
1980 1990 2000
Above: Figure 4 – Fast Fourier Transform based curve fitting (show in blue) to averaged Zambia – Zimbabwe
Reconstructed Rainfall data since 1800 (shown in red); Figure 5 – Cumulative Discharges in major African
from 1945 to 2000 divided by their mean flows for the same period
WWW.WATERPOWERMAGAZINE.COM OCTOBER 2010 15
INSIGHT
marises periodicities from examples in the
Middle East, China, North America, Africa
and northern Europe.
Various groupings of periodicity are evident
in Table 1. Those of between 3 and 3.8 years
almost certainly correspond to El Nino. Those
of 18.6 years could correspond to that of
known lunar tidal effects. It has been argued
that those close to 11, 22, 90 and 200 years
match known periodicities in solar variation.
Lessons for the future
Cyclic variability in weather patterns has been
explored and debated in the past but for practi-
cal engineering purposes it has generally been
ignored. Conventional analyses for hydropow-
er yield and ood assessment generally assume
any records to be random. One problem with
periodicities is that unless one is actively look-
ing for them one is unlikely to nd them. Cyclic
trends are often intermittent and discontinuous
while their causal mechanisms are also often
unclear and without that understanding there
is natural caution against relying on them.
However with increased availability of data
and the increased ease with which such data
can be processed surely it is about time the pro-
fession expanded its horizons. As demonstrat-
ed in the preceding sections, an understanding
of periodicity in the record can improve invest-
ment timing. It can also improve an under-
standing of required flood design capacity.
The assessment of design oods based on a
hydrological record is meaningless unless one
knows whether the basic data set corresponds
to a dry, average or wet phase in terms of long
term hydrology. This is not only true for nal
works but also for planning the capacity and
timing of river diversion during construction.
In the wider context an improved understand-
ing of apparent periodicities in the natural record
would seem to offer at least one planning sce-
nario to be considered in terms of investment
and even for the long term planning of aid and
famine relief. It can only be hoped that as cli-
matic data sets continue to be expanded, their
analyses will include not only statistical prob-
abilities but also cyclic variability as an aid to the
long term planning of those aspects and also of
major civil infrastructure.
Dr Peter Mason can be contacted at
peter.j.mason@mwhglobal.com
IWP& DC
2
1
0
-1
-2
1100.0
900.0
700.0
500.0
300.0
100.0
–100.0
30.0
20.0
10.0
0.0
-10.0
-20.0
-30.0
2
0
2
19401920 1980 20001960
Year
0
0
0
0
0
0
0
3
2
1
0
-
1
-
2
-
3
1881
1885
1889
1893
1897
1901
1905
1909
1913
1917
1921
1925
1929
1933
1937
1941
1945
1949
1953
1957
1961
1965
1969
1973
1977
1981
1985
1989
1993
1997
2001
2005
Year
Precipitation in mm
Annual precipitation Cumulative departure
fromMAP (scale x 10)
Sunspot number:
Cumulative departure from mean
Dublin: Annual precipitation 1881-2006
2
1
0
-1
-2
1100.0
900.0
700.0
500.0
300.0
100.0
–100.0
30.0
20.0
10.0
0.0
-10.0
-20.0
-30.0
2
0
2
19401920 1980 20001960
Year
0
0
0
0
0
0
0
3
2
1
0
-
1
-
2
-
3
1881
1885
1889
1893
1897
1901
1905
1909
1913
1917
1921
1925
1929
1933
1937
1941
1945
1949
1953
1957
1961
1965
1969
1973
1977
1981
1985
1989
1993
1997
2001
2005
Year
Precipitation in mm
Annual precipitation Cumulative departure
fromMAP (scale x 10)
Sunspot number:
Cumulative departure from mean
Dublin: Annual precipitation 1881-2006
References
Alexander, W. (2007). Locally Developed Climate Model Verified. The Water
Wheel, South Africa, Jan/Feb.
Burroughs, W.J. (2003). Weather Cycles Real or Imaginary? Second Edition,
Cambridge University Press, UK.
Brooks C. E. P. (1926). Climate Change Through the Ages, Ernest Benn Ltd.
London, 107-136.
Brooks C.E.P. and Carruthers N. (1953) Handbook of Statistical Methods in
Hydrology, HMSO, London, 1953, p 330.
ESRL Quarterly, Fall 2008. (2008). National Oceanographic & Atmospheric
Administration (NOAA), Boulder, Colorado, p5.
Frossard E, et al. (2006). Evolution de Regimes Hydrologiques avec le Climat:
Incidences sur les Amenagements Hydrauliques, International Congress on
Large Dams, Barcelona, 2006.
Hurst H.E. (1952). The Nile, Constable, London, 263-269.
Kafue Gorge Lower Hydroelectric Project, Climate Change Impact Draft
Report (2009). Prepared for the World Bank (IFC) by MWH.
Kite G.W. (1982). Analysis of Lake Victoria Levels, Hydrological Sciences
Journal, 27(2), 6, 99-100.
Lamb H.H. (1966). Climate in the 1960s, The Geographical Journal, 132,2,
June, 183–212.
Mason P. J. (1993). Hydrology or Climate Change, a Case Study, Dams and
Reservoirs, The Journal of the British Dam Society, 3(2), June, 10-12.
Mason P.J. (1998). Sunshine over Lake Victoria, The Chemical Engineer, 6
August, 12-14.
Mason P. J. (2006a). Lake Victoria: a Predictably Fluctuating Resource,
Hydropower & Dams, Vol. 13, 3, 118-120.
Mason P. J. (2006b). Lake Victoria: a Predictably Fluctuating Resource,
Hydro 2006 Conference, Porto Carras, Greece
Mason P. J. (2010) Climate variability in civil infrastructure planning,
Proceedings of ICE, Civil Engineering 163, May 2010, 74-80.
Mauas P. and Flamenco E, (2005) Solar activity and the stream flow of the
Parana River, Memoria de la Societa Astronomica Italiana, Vol. 76, 1002.
Mauas P., Flamenco E and Buccino A.P., (2008) Solar Forcing of the Stream
Flow of a Continental Scale South American River, Physical Review Letter
101,168501, 17 October 2008, The American Physical Society.
May B. R. and Hitch T. J. (1989). Periodic Variations in Extreme Hourly Rainfalls in
the United Kingdom, The Meorological Magazine, Mar, 118, No 1400, 45-50.
Ruzamaikin A. et al. (2006). Is solar variability reflected in the Nile River,
Journal of Geophysical Research, 111, D21114.
Rydgren B. et al. (2007). Addressing Climate Change-Driven Increased
Hydrological Variability in Environmental Assessments for Hydropower
Projects – a Scoping Study, The World Bank, Washington, June.
Stager C. et al. (2007). Sunspots, El Nino and the levels of Lake Victoria,
East Africa, Journal of Geophysical Research, 112, D15106.
Therrell M.D. et al. (2006a). Zimbabwe Summer Precipitation Reconstruction
1796-1996. IGBP PAGES/World Data Center for Paleoclimatology, Data
Contribution Series # 2006-057. NOAA/NCDC Paleoclimatology Program,
Boulder CO, USA.
Therrell M.D. et al. (2006b). Tree-ring Reconstructed rainfall Variability in
Zimbabwe. Climate Dynamics, 6, 677-685.
Above: Figure 6 – De-trended time series for Parana flows (in black) and similarly de-trended sunspot numbers (in red). Both series were normalised by subtract-
ing the means and dividing by the standard deviations; Figure 7 – Comparison between the cumulative departure of annual rainfall recorded in Dublin and the
cumulative departure of annual sunspot numbers
16 OCTOBER 2010 INTERNATIONAL WATER POWER & DAM CONSTRUCTION
CONSTRUCTION
H
YDROPOWER currently provides just 40% of the
UK’s renewable energy production, but changes in gov-
ernment policy have provided opportunities to develop
hydropower projects which previously would have been
unviable. The UK, which already has over 1500MW of installed
hydropower capacity, generating over 4000GWh/yr, is now much
better placed to make use of the water resource available.
Recent surveys have shown that there is potential in the UK for a
further 2GW of capacity, with signicant further development pos-
sible, particularly for new micro scale hydropower projects. Many
of these micro generation projects will make use of existing infra-
structures such as old mill sites and weirs, or water utility assets such
as reservoirs and water transfer schemes, as well as opportunities to
enhance existing hydro schemes.
MICROGENERATION
In April 2010, the feed-in tariff (FIT) scheme was introduced, incen-
tivising small scale (less than 5MW) low carbon electricity generation
and making low head schemes a commercially viable option.
JN Bentley, a well established civil engineering contractor, has been
quietly building its reputation in the hydropower sector for a number
of years. Through the company’s collaborative working agreement
with Archimedean screw supplier Spaans Babcock, it is now using
its expertise to install screw powered hydro generators on a range of
low head sites. Many more schemes are in the planning and outline
development stages.
Closest to its heart is JN Bentley’s own hydro development site at
Linton Falls in the beautiful Yorkshire Dales National Park, only a
few miles from the head ofce in Skipton.
The Linton Falls site was originally built as a hydropower station
and operated from the early 1900s to 1946 when the National Grid
nally brought power to the Dales. Such is the importance of the
site, as an exceptionally rare example of early industrialisation in the
Yorkshire Dales, that what remains of the now derelict powerhouse
is listed as a Scheduled Monument by English Heritage.
JN Bentley has received consent from English Heritage to sym-
pathetically restore the 100-year-old crumbling turbine house to
UK renewable energy legislation has driven
innovative new applications of hydropower
technology. Samantha Wood from civil
engineering contractor JN Bentley reports
Right: View of Linton Falls
Below: Section of the Linton Falls turbine
A quiet revolution
WWW.WATERPOWERMAGAZINE.COM OCTOBER 2010 17
CONSTRUCTION
its former use and secured planning permission and Environment
Agency licensing to build and operate a 100kW twin Archimedean
screw hydro station.
By 2011, Linton Falls will be generating sustainable energy for
the local communities once again, this time via the National Grid.
This project is an example of a growing trend in the UK, with the
Yorkshire Dales National Park Authority alone considering up to 50
further sites for small hydro schemes.
LARGE SCALE RUN-OF-RIVER
The skills developed during the design and development of Linton
Falls have been taken forward by JN Bentley’s team of hydro special-
ists to help clients realise similar aspirations. Many of these are also
weir based projects, but also include Kaplan turbine technology in
addition to Archimedes screws.
JN Bentley is currently working with the Small Hydro Company,
in partnership with British Waterways, to develop run-of-river hydro
schemes on around 25 sites across the organisation’s 3540km water-
way network. The wider project team also includes design consultants
Mott MacDonald, sheries experts Turnpenny Horseld Associates
and JN Bentley’s supply chain.
Early contractor involvement (ECI) during site appraisal and
scheme development stages is ensuring that ‘buildability’ and com-
mercial viability do not become issues, by quickly discounting unsuit-
able sites and minimising the cost of design development.
The 25 proposed hydro schemes, ranging in electrical capacity from
100kW to 1.5MW, will generate 210,000MWh of renewable energy
a year, enough for about 40,000 homes. Gunthorpe and Sawley are
two key sites currently under development which combined will con-
tribute 33% of government targets for hydropower generation in the
East Midlands region.
The Sawley Weir site, on the River Trent near Nottingham, will
be the rst of the two sites to be completed and will use two Kaplan
turbines at the weir to generate up to 1MW of power. Sawley Weir,
under normal ow conditions, is around 1.8m high, but the high ow
rate of the River Trent is best used by increasing the head and using
twin 2.5m diameter Kaplan turbines.
While increasing the head makes the site more commercially viable
it also brings a number of challenges. The biggest issue was public
perception of an increased ood risk.
In proposing an increase in weir height by 800mm, there was an
immediate concern that the water level upstream of the installa-
tion would also increase by 800mm and therefore increase the risk
of ooding. The challenge was in reassuring those concerned that
Kaplan turbines of this size actually have the effect of drawing down
the water level upstream, in this case by approximately 600mm.
Such a small net increase in water level is well within the natu-
ral variation of water level on the river, but to prove this required
extensive use of a computational model of the River Trent, obtained
from the Environment Agency, combined with site surveys and tra-
ditional calculation.
Once the team had overcome this, the next concern was the potential
for the turbines to have to shut down due to mechanical/electrical issues
or the weir ooding out and the head disappearing during oods.
Unchecked, this would exacerbate upstream water levels as the
draw down effect would be lost. But to prevent this situation, the
hydro team incorporated twin hydraulic ood gates in the design
which will open and release trapped water from upstream of the
installation. When open, the gates will allow the same volume of
water through as the increased weir height had prevented, making
the installation ood neutral and maintaining the river’s natural ood
response.
JN Bentley’s ECI during design development, together with the
ood assessment expertise of consultant Mott MacDonald, both over-
came public concerns and maximised the power generation potential
Linton Falls, Yorkshire
Head: 2.8m
Max flow rate: 4.5m
3
/sec
Installed capacity: 100kW
Annual production: 516MWh/yr
Homes powered: 122
*1
Annual CO2 saved: 258 tonnes
*2
Sawley
Head: 2.2m
Max flow rate: 45m
3
/sec
Installed capacity: 1MW
Annual production: 100MWh/yr
Homes powered: 1700
*1
Annual CO2 saved: 3550 tonnes
*2
Gunthorpe
Head: 2.5m
Max flow rate: 68m
3
/sec
Installed capacity: 1.2MW
Annual Production: 8800MWh/yr
Homes powered: 2100
*1
Annual CO2 saved: 4400 tonnes
*2
*1 Homes powered based on typical 3 bed room house annual use of
4200kWhr/yr
*2 CO2 saving based on typical 0.5kg per KWhr generated by fossil fuels
typically in UK.
Above: Typical section of the turbine scheme at Gunthorpe and Sawley
18 OCTOBER 2010 INTERNATIONAL WATER POWER & DAM CONSTRUCTION
CONSTRUCTION
of the river. As a result, the scheme achieved planning consent within
two years of scheme conception, an exceptionally short timescale for
a UK hydro scheme of this size.
The scheme at Sawley has also presented the opportunity for addi-
tional environmental enhancement. The Small Hydro Company has
made a commitment to install sh passes on all their hydro schemes,
regardless of location or type of installation; in this case, supporting
the objectives of the Environment Agency and the Trent River Trust
to re-establish a sustainable salmon population in the River Trent
by 2020.
JN Bentley has also secured contracts with RWE npower renewa-
bles. The £4M (US$6M) civils contract for the Black Rock run-of-
river scheme in the northeast of Scotland involves a 3.5MW twin
Francis turbine installation on the River Glass at Black Rock Gorge
near Inverness. It also includes construction of a powerhouse and low
prole concrete intake weir. The project utilises a 70m fall in river
level between the new intake weir and powerhouse, with approxi-
mately 3.5km of buried 1.5m diameter pipeline installed to connect
the two. The scheme is due for completion in autumn 2011 and will
supply power to the National Grid.
The 47-week project will produce an average of more than 10GW
hours each year – enough to supply around 2000 households with
their annual electricity needs.
VALUE-ADD HYDRO APPLICATIONS
The same hydropower technology is also proving to be a signicant
value add-on for some major projects where power generation is not
the primary purpose, making an important contribution to opera-
tional efciencies.
In February 2007, in the rst installation in the UK to use untreated
sewage for hydropower generation, JN Bentley installed two new
Archimedean screw hydro-turbine generators at Yorkshire Water’s
Esholt wastewater treatment works as part of an overall site upgrade
worth £19M (US$29M).
The two generators supplied by Spaans Babcock, each weighing
more than 20 tonnes, were installed in series between the inlet works
and the new primary settlement tanks. Up to 3240L/sec of waste-
water now ows sequentially through the two screws over a total
fall of around 10m, passing through an 1800mm diameter pipe to
the hydro turbine station. Over 180kW is generated, reducing the
imported power demand of the treatment works.
Another agship UK project is the £4.6M (US$7M) redevelop-
ment of the Tees barrage white water course in Stockton on Tees
in northeast England, funded by regional development agency One
North East, Stockton Borough Council, Sport England and British
Waterways. The project is currently the only installation in the world
where Archimedean screws are used as both pumps and generators.
A number of improvements are underway to turn the course into a
world-class water sports facility and it has already been conrmed as
one of the ofcial training camps ahead of the London 2012 Olympic
and Paralympic Games.
Improvements will include reconguring the existing main course
(a 250m long, 7m wide ‘U’ shaped loop, with a 2.5m drop and a
ow of 10m
3
/sec) and building a second shorter, steeper course for
high-speed paddling.
Before redevelopment, the course could only operate two or three
hours either side of low tide in the River Tees. To extend the operat-
ing time, and improve course ow control, JN Bentley is installing
four 12m long by 3.1m diameter Archimedean Screws, each weighing
more than 30 tonnes, in a newly built pumping station.
The screws will pump water from the bottom pool to the top pool,
creating guaranteed conditions for canoeists and rafters regardless
of tide. When not pumping water around the course, the screws will
turn into generators using excess river water to generate electricity
for the National Grid, making Tees the UK’s rst fully sustainable
white water course.
When operating as a screw pump, the power absorbed at the duty
point (3500 l/sec @ 4.94 m) will be 178.5 kW per screw pump. The
screws will each be capable of lifting 3.5m
3
/sec at a maximum head of
3.68m, producing a maximum ow rate of 14m
3
/sec for the course.
The screws are controlled by variable speed drives via a SCADA
controls system in the main control tower, and can be run in either
manual or automatic mode.
In manual mode the pumps will run at a xed speed of 1500rpm
from an input of 20mA. When the pump is running up to speed, the
system will send signals to the hydraulic power pack to open the rel-
evant Penstock. Automatic mode will control the number of pumps
needed from those available and vary their speeds to keep the bottom
pool at its desired level of -1m. If more than one pump is available,
the system will only vary the speed of one pump to trim the level
and run the others at their most economical speed. It will also select
available pumps according to their accumulated hours to balance the
running time as far as possible.
Generating mode will only be available when pumping mode is set
to ‘off’. When operating in reverse, generating electricity, each screw
pump will produce approximately 131kW. To maximise the oppor-
tunities for exporting electricity to the National Grid, the system will
automatically alert the barrage operator when the conditions are
appropriate i.e. no pumps in ‘pumping mode’, upstream level above
2.35m AOD and lower pool level below -1m.
The project, being led by British Waterways, is due for completion
in winter 2010.
THE FUTURE
As populations grow and carbon reduction targets are set, the increas-
ing need for sustainable energy is one of the few certainties the civil
engineering and construction industry can expect over the coming
years. Designers, contractors and clients alike, who are positioning
themselves to address tomorrow’s needs today, will be well placed to
be leaders when this still youthful market reaches maturity. The revo-
lution may be quiet, but it’s coming.
Samantha Wood, Communications Ofcer, JN Bentley.
The author would like to thank the following at JN
Bentley for their help and technical expertise in producing
this article: Austin Flather, Rhianna Rose, Kate Crawford,
Oliver Wilson, Graham Jessop and Paula Wyldbore.
Permission kindly granted by: British Waterways, RWE
npower renewable, Small Hydro Company, Spaans Babcock,
Turnpenny Horseld Associates and Yorkshire Water.
IWP& DC
Above: Work in progress at the Tees barrage; Tees barrage 3D CAD
!"##$%& '
! "" # $ %#$
&' ()*%*' +,--.&)
% *, /0+ ') 1 -
& ()*+
&' *%
233 . 4
"$ 1562675 8$ 15626756
$99"$9 : ;99"$9
$$$%! #"
!
Untitled-1 1 13/7/10 09:00:27
!
"#$ $$%&'$$%&(
)* +& , ,&
-./0 1/2"0) ).-"/ +/1"
20 OCTOBER 2010 INTERNATIONAL WATER POWER & DAM CONSTRUCTION
CONSTRUCTION
L
OCATED on the middle reaches of the Jinsha River,
Jinanqiao hydropower plant is the fth in a series of eight
hydroelectric power stations. The dam site is located 50km
away from Lijiang City in Yunnan Province in China, and
has a catchment area of approximately 237,400km
2
. The average
inow at the dam is 1640m
3
/sec, annual runoff is about 51.7km
3
,
normal reservoir water level is 1418m, with a corresponding stor-
age capacity of 847Mm
3
, including 313Mm
3
of regulated capac-
ity. The powerhouse is located downstream of the dam, with an
installed capacity of 2400MW (4 × 600MW).
The project comprises a dam, power plant, crest spillway, and
stilling basin on the right bank, two bottom outlets for sand sluicing
and ood discharge on the right of the dam, and the third bottom
outlet for sand sluicing on the left of the dam.
The dam is a roller compacted concrete gravity dam with a crest
elevation of 1424.00m; the maximum dam height is 160m with a
crest length of 640m and downstream slope of 1:0.75. The upstream
slope below elevation 1335.00m is 1:0.3, and the concrete volume
of the dam is 3.92 × 10
6
m
3
. Five openings at the crest with a size of
13m × 20m form the spillway; its discharge capacity is 14980m
3
/sec
(5000 years design ood). There are four penstocks, one for each
hydroelectric unit. The diameter of each penstock is 10.5m, ow
is 605m
3
/sec, and the rated head is 111m. There are four access
sluices whose dimensions are 9m × 14m, as well as four emergency
gates with dimensions of 9m × 12m in the power station intake. The
dimensions of the powerhouse are 213m × 34m × 79.2m (length
× width × height); four sets of 600MW Francis turbine units with
runner diameters of 8m are installed within the power plant.
According to research results from the Institute of Geology,
State Seismological Bureau of China, the basic seismic intensity of
the dam site is 8 degrees on the Chinese intensity scale. The peak
horizontal ground acceleration of the bedrock site for earthquake
with return periods of 5000 and 10,000 years is 0.4 g and 0.475 g,
respectively.
This project was approved in August 2003; and the river was
closed at the end of December 2005 (Figure 1). The main construc-
tion works began in January 2006, with the dam foundation exca-
vation completed and the rst concrete poured in February 2007
(Figure 2). The project’s rst generator went into operation in June
2010.
DESIGN AND CONSTRUCTION MEASURES
Integration of dam body and powerhouse to improve sliding stability
There is still a lack of information on the special seismic design fea-
tures of large RCC gravity dams in areas of high seismicity. Under
seismic action high dynamic tensile stresses occur in the dam body,
especially at locations of geometrical changes. In addition, the sliding
stability of the dam has to be guaranteed. In Figure 4 two sliding fail-
ure modes for the dam foundation are shown, which account for the
special geotechnical features in the dam foundation. Two structural
systems are shown where (a) the powerhouse and the dam body are
separated, and (b) where the dam and the powerhouse are acting as
a monolithic structure by grouting the joint between the dam and the
powerhouse.
In system (b) the sliding stability can be improved considerably
(Yuan, 2008).
The incomplete cut-off of the transverse joints
Because of the special features of the foundation rock, the sliding
stability of the left bank concrete blocks under seismic action is of
major concern.
In order to improve the sliding stability of the dam, and to mobilize
the whole mass of the dam, the depth of the contraction joint is
only 2/3 of the dam thickness in each RCC layer (each RCC layer
is 30cm deep). The electronic power cutting machine, with cutting
blade width of 1cm, is used to generate the joint at the presetting
transverse joints after each layer is compacted by the roller. The
depth of the cut-off is 20cm, which approaches two-thirds of the
thickness of the layer. The 10cm thick RCC, without cutting in
the layer, still interconnects. Non-woven fabrics are embedded in
Yong-Wen Hong and Cheng-Bin Du present details on the design and construction of the
Jinanqiao RCC gravity dam, a 160m high structure located in China’s Yunnan Province, in
a region of high seismicity
Building Jinanqiao Dam
Figure 1: Jinsha River closure in 2005 (17 December, 2005) Figure 2: Concrete placement (28 February, 2007)