Dams Part3
Dams Part3
For climates with less intensive rainfalls such as lower, coastal African locations
and some North American and Australian environments, the peak floods would be
lower and the spillways that much smaller.
160
140
120
100
80
60
40
20
0
5 10 20 50
RETURN PERIOD (YEARS)
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
1 2 3 4 5 6 7
Tc (HOURS)
0.2
0.1
0
5 10 15
EXTREME HEIGHT SLOPE (%)
Where data are available, it would not be difficult to draw up similar charts or
tables and once peak floods are determined, the hydraulic parameters for estimat-
ing spillway widths and depths are available.
In all cases, however, such charts and tables are by their nature generalized and
should always be used with caution and, wherever possible, be adapted to suit local
conditions.
5 Site selection and preliminary investigations 39
Once the PMF has been estimated, the spillway width can be calculated using the
formula:
Qp = 1.7 b D1.5
‘1.7’ is a factor derived for concrete ogee type crests and can vary up to 2.25 accord-
ing to site conditions and factors of safety. 1.7 is generally used for spillways for
small dams on small catchments.
‘b’ is the minimum width (‘breadth’) of the spillway and is calculated by introduc-
ing the values for Qp (estimated using the options above) and D1.5. It is assumed
that b is large when compared to D and that the spillway channel will thus be
rectangular.
‘D’ is the depth of the spillway at the crest and will comprise all or part of the
design freeboard. D is normally in the range 0.75 m to 1.5 m for small dams and
comprises the total freeboard. However, where wave action or backing up of floods
may affect the dam, an additional ‘dry’ freeboard of up to 0.75 m. should be added
to the figure above for safety reasons.
Once all the other values are known, ‘b’ can then be calculated and the best option
for varying depths, ‘D’, can be chosen.
The width ‘b’ is the minimum width for the spillway to accommodate the design
flood. It assumes that there is no constriction downstream of the spillway. The
width and depth may have to be adjusted to suit the local topography and spillway
bed material later in the design process.
This will, for irrigation dams, comprise irrigation requirement, other uses (live-
stock/domestic water), losses to seepage and evaporation and dead storage.
¼ Irrigation requirement can be calculated by multiplying the gross annual
irrigation requirement per hectare by the area proposed. This may have to be
adjusted once the estimated storage for the dam chosen is calculated.
¼ Environmental flows to release normal flows into the river or to comply with
any legal requirements downstream.
¼ Other uses such as livestock water can be calculated by estimating water use
for this. FAO can provide advice as well as locally based government and
other organizations. As a guideline the following (assuming the animals are
on dry pastures and good quality water is available) can be used:
– Cattle 40-80 litres/day for each animal (milking cows may
need 100 litres/day).
– Young stock 25-50 litres/day.
– Pigs 25 litres/day.
– Poultry 30 litres/day per 100 adult birds.
– Bee hive 2 litres/day.
– Sheep 2-6 litres/day.
– Goats 3-8 litres/day.
40 Manual on small earth dams
Add 10 percent to any calculated total for water use by wild and feral
animals and add a further 10 percent if the water is higher in salt content than
recommended. Slightly saline waters can be tolerated by animals (but pigs
and poultry are most sensitive) but they will have higher intakes to allow a
greater water turnover to regulate body salt balances.
¼ Troughs are always recommended. Dams should be fenced off and no
livestock allowed to drink directly from the reservoir or to damage the
surroundings to the dam by overgrazing the catchment, tracking in the
immediate surrounds of the reservoir and wallowing in the reservoir itself.
¼Domestic water uses – opting for piped water supplies using filters or similar
– can be calculated by determining the likely numbers of people who will
use the dam for water and estimating total annual or dry season needs. A
minimum of 20-50 litres/day per person in more rural areas can be used if
piped water supplies are not to be provided but consideration for increases
in use should be made in areas where populations are high and levels of
urbanization may increase.
¼Seepage losses are always difficult to estimate before the dam is built and to
calculate after the dam has been constructed. As all dams will seep, it is best
to estimate that a well constructed embankment will lose about 10 percent of
its water to seepage in any one year.
¼Evaporation losses can be calculated from local records noting that shallow
large surface area reservoirs will have higher evaporation rates than narrow
deep reservoirs. Wind is also an important factor in dry areas. Annual rates
of evaporation from dams in Africa can exceed 30 percent but for calculating
water uses (i.e. for irrigation), where actual figures are not known, dry season
losses can be taken as 20 percent maximum.
¼Dead storage is the amount of water retained in the dam that cannot be
accessed. The dead storage will vary according to design, pumping suction
heads and positions of any outlets in the embankment. It will also be more,
proportionally, for a small dam than a larger dam and will offer an area in all
dams for sediment to accumulate. For design purposes, a figure of 5 percent
maximum of the total water stored can be used to estimate dead storage.
Once the above has been estimated the remaining amount available for irrigation
can be calculated. It is at this stage the areas proposed under irrigation can be
adjusted and any economic analysis made.
Detailed design 6
6 Detailed design 43
6 Detailed design
6.1 INTRODUCTION
Once all preliminary investigations have been made and a suitable site has been
found, the next step is to carry out a detailed survey of the valley and reservoir area
to allow more accurate estimates of quantities and to provide the necessary data for
design work to be undertaken. The aim of such a survey is to present, on paper, a
contour map of the reservoir up to and exceeding the maximum flood level, and
to provide details for the location of the embankment, spillway and outlet works.
From the contour map, the capacity of the reservoir can be assessed for varying
dam heights. A depth-capacity curve can then be drawn up to provide a quick
and easy method for the dam designer to choose the optimum full supply level.
A simplified example of a depth-capacity curve is shown in Figure 7. Often the
depth-surface area curve (usually with a reversed scale) is added to these graphs.
3.0
2.0
1.0
0
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
14 High levels of accuracy are not required at this stage considering that elevations will later be
affected by site stripping pre-construction.
44 Manual on small earth dams
1. Grid survey
This is a simple and straightforward but time-consuming method. Also it may not
be possible if the area is heavily vegetated and/or physically inaccessible.
2. Cross-sections
Cross-section surveys are taken along various lines within the river valley(s)
from benchmarks previously established. Levels are observed at set intervals and
outstanding features (changes of slope in particular) are also noted.
3. Spot heights
This is especially suited to larger areas. A circuit of benchmarks is established and
spot height observations with bearing, distance and elevation are made from each
station. For smaller dams, and if a theodolite or electronic instrument is used, it
may be possible to take all the readings from one station. Alternatively, reason-
ably accurate GPS surveys can be used to establish a network of elevation readings
across the site.
Q= H’ A’
3
Fill in the reduced level column on the left-hand side of the sectional paper, starting
with the settled crest level on the top line. It is advisable, for ease of working, to
consistently use a reference reduced level of 100 (largely to avoid having negative
6 Detailed design 45
values when referring to the crest height and to make any calculations above or
below this reference level easy to work out) either for the highest or for the lowest
point of the proposed embankment.
Draw the cross-section of the proposed dam at its maximum height (i.e. above
stream bed) after settlement, starting with the upstream toe on the left at zero
(using the horizontal scale at the bottom and marked base width), working up to
the crest, along and down to the downstream toe. This plotting must be carried out
accurately as scaled dimensions are to be used in the calculations. Calculate, and
check by measurement, values of w (i.e. the mean width of each 0.5 m or 1 m cross
section) commencing with the crest section and enter it in the appropriate column
and line.
46 Manual on small earth dams
Measure carefully the values of l (i.e. the length of the longitudinal section to corre-
spond with each position of w) and again enter it into the appropriate column and
line. Multiply each w by the corresponding l, and the height of the section h, and
enter the result in the Volume column on the appropriate line. Total this column to
give the volume of the earthworks in the dam.
1x29.5
105 29.5 106 3 127
5
1x33.0
104 33.0 63 2 079
4
1x36.5
103 36.5 31 1 131
3
1x40.0
102 40.0 11 440
2
1x43.5
43.5 9 392
101 1
1x47.0
47.0 25 118
100 0
0 5 10 15 20 25 30 35 40 45 48.5
WIDTH w (m) EMBANKMENT VOLUME 28 417 m3
TR. BANK VOLUME 294 m3
TRAINING BANK
V = hl (c+ hs) m3 where h = mean height = 0.9 m TOTAL VOLUME 28 711m3
VOLUME 2
l = length = 75 m SETTLEMENT 10% 2 871 m3
c = crest width =2m
FINAL VOLUME 31 582 m3
s = slopes (1.5+1.5) =3m
Spillway volume excluded
V = 0.9x75 (3+ 0.9x3) = 293.65 m3 w is mean width of x-section
2 s is total value of both side slopes
Comparison of this result with the result obtained from the formula
is likely to lead to quite a significant difference. The formula result should be much
greater, but this is acceptable as the formula is used for calculating the total cost
of the dam and not the earthworks alone. At the reconnaissance stage, a general
planning cost is required and this is calculated by using an estimated overall rate
per m3 of earthwork that will include all of the items detailed in Table 3 in the next
section.
6 Detailed design 47
Method 2
Method 2 (see Figure 8b) is much the same as Method 1 except that the cross-
sectional area is calculated more accurately. The cross-section is squared off as
illustrated and each rectangle has its respective area calculated in a straightforward
manner (i.e. length x breadth). The remaining triangular pieces which flank each
rectangle have constant areas that are calculated as follows:
¼ upstream slope 1:2, height of section 1 m,
area of upstream triangle= (2/2) x 1 = 1 m2
¼ downstream slope 1:1.75, height of section 1 m, area of downstream triangle
= (1.75/2) x 1 = 0.875 m2
Therefore, each cross-sectional area can now be estimated relatively quickly and
the method of assessing volumes proceeds as in Method 1.
Finished versions of the method can then be presented on design drawings with
allowances for over-excavation, training walls and settlement and without the
calculations.
location is essential. Further drawings for more specialized aspects of the works
can also be provided.
Should the dam (or dams) design and costing be prepared for tender or contracting
to the private sector it is important that the details on costing for Table 3 and any
engineer’s estimates remain confidential and be used as a guide in evaluating any
bids or other proposals from potential contractors to construct the dam(s). Annex
1 has more details on this.
Critical items are the entrance width ‘b’ (already discussed above and dependent
on the peak flood), the outfall (dependent usually on ‘b’ – refer below) and the
material the spillway will be constructed with and sited upon.
Guidelines to follow are presented in Table 4. The guidelines assume that an earth
spillway is level and grassed with good, mat-forming creeping grass.
The main spillway can be reserved for flood flows and problems of gullying will
thus be avoided (erosion may still occur as a result of floods. This should only
happen infrequently and can be treated accordingly).
The pipe, as illustrated (with inlet chamber) in Figures 9a and 9b, must be carefully
laid, true to line and level. Steel pipes should be flanged and concrete pipes should
have staunching rings (anti-seepage collars) to prevent seepage of water along the
50 Manual on small earth dams
outside of the pipe. The pipe should be laid in a trench cut in original ground on
the valley sides before the embankment is built. If stream flows are not known, the
minimum diameters of pipe are as follows:
– 300 mm for very small catchments.
– 375 mm for catchments up to 5 km2.
– 450 to 550 mm for catchments between 5 and 8 km2 (i.e. ‘44 gallon’ drums in
concrete).
For known expected maximum stream flows the diameters of the pipe and its
physical gradient can be selected from Tables 5a or 5b.
Grass, and occasionally rock spillways, may require the construction of training
banks (stone pitched if necessary) to guide the flood flows away from the steeper
52 Manual on small earth dams
slopes and the downstream toe of the dam. A maximum slope of about 5 percent
for the return should be the goal and this should only be exceeded where rock is
to be used for the return. The actual outfall should be designed to be non-erosive
and, as a rule of thumb, the final width should be 1.5 to 2 times the entrance width
‘b’ thus reducing velocities of flow to manageable levels. Examples of training bank
and outfall designs are provided in the sample layout drawings in Annex 4.
On rock spillways, downstream erosion that will not endanger the embankment or
cause environmental problems and which will stop once the flow has eroded back
to the rock, is permissible. For drop inlet overflows the construction of a channel
of brick or stone from pipe outlet should be sufficient and this can then be led to a
safe dissipation point downstream.
If farm machinery or other vehicles are expected to use the embankment and
spillway as a road, the side slopes of the spillway should not exceed 25 percent and
some protection (i.e. stone or concrete crossings) from erosion by traffic should be
constructed at the time of building the dam.
If the dam is already constructed and an outlet pipe is required, excavation into
the embankment is not recommended as this would create an area of weakness in
what is meant to be a unified structure. The alternatives are either to pump from
the upstream side over the embankment or to construct a siphon.
The pump(s) could be located on a raft with a flexible connection to a fixed pipe
on the dam, or be positioned on a ramp that will allow them to follow water levels
as they rise and fall to avoid too high suction lifts occurring (i.e. more than 3-5 m).
Siphons require careful construction to ensure all joints and valves are airtight and,
as insurance, some means of priming at the highest point may be incorporated in
the pipeline. With a siphon it is essential that the outlet be located at a level below
that of the inlet when the water level in the reservoir is at its lowest. Siphoning
water over an elevation of more than 5 m is not advisable and it may be necessary
to minimize the elevation difference by burying the pipe into the top portion of
the embankment.
wave action, the slopes should be made flatter to suit the circumstances involved.
Anthills and solid rock outcrops should be avoided unless there is no alternative.
Anthills should be completely excavated and the hole filled in, preferably with soil,
or, as a last resort, with treated ant heap material in well compacted thin layers.
Rock outcrops will require scraping down and key walls built into the embank-
ment or core.
6.9 FREEBOARD
Freeboard for small dams should never be less than 0.5 m with 0.75 m to 1.0 m
preferred. Where wave action is likely, additional freeboard may be required. This
can be estimated using the following formula:
where H’’ is the freeboard height and F is the fetch which is the longest distance,
in km, across the storage area (usually measured in a straight line from the centre
line of the proposed embankment to the tailwater area of the proposed reservoir).
The overall freeboard height can then be calculated taking into account the wet
freeboard, H’’, (as estimated with the formula above) required to counteract wave
action and the dry freeboard (estimated by the engineer) for safety and other
factors. The total freeboard is effectively the design depth for the spillway (at its
entrance).
For designing small dams, as the cutoff can be excavated by hand or small machin-
ery, it need not exceed 2 m wide. For larger dams, cutoffs can be excavated by
bulldozer or scraper and then will require a width, usually 4 m, that permits access.
Excavation of any trench requires safety factors to be considered and, for deep
trenches, benched or sloping sides or other measures may be required to reduce
the possibilities of the sides collapsing. Sloping or benched sides also permit easier
compaction and improves the bond between the backfill and the existing ground.
54 Manual on small earth dams
Cw (in m) = 0.4H + 1
where Cw is the crest width and H is the maximum height of the dam in metres.
Always adopt the widest crest width possible (and flatter embankment slopes)
where foundations or construction materials are suspect.
To reduce erosion, all crests should be given a 2.5 percent crossfall to drain rainwa-
ter to the reservoir via the upstream slope of the embankment.
6.14 SEEPAGE
Seepage is always a potential problem that should be considered at this stage and
the designer-builder will have to bear in mind the permeability of the fill materials
and of the foundation, the position and flow of groundwater at the site, the type
and design of any core or below ground cutoff within the embankment, and the use
of drainage devices to collect and safely channel seepage water in the downstream
section of the embankment. All earth dams will have some seepage and it is unre-
alistic not to expect this. If seepage is considered as a potential problem, counter-
measures – such as filters, drains, clay blankets and flatter side slopes – introduced
at the design stage can reduce any risks to a minimum.
6 Detailed design 55
The aim of all seepage ‘filter’ drains is to lower the phreatic surface (the ‘seepage
line’) within the embankment to prevent water from emerging from the down-
stream slope where erosive and absorptive flows could cause slumping of the
material and endanger the whole structure.
Trenches dug into the subsoil beneath the downstream face and toe, at the time
of construction, and filled with rock and gravel (the latter helping to limit the
movement of finer embankment material into the drains) and continued to a collec-
tor drain network at least 3-5 m below the toe line, can safely bring seepage lines
down to allow flow out from beneath the embankment.
The configuration of the filter zones, however, will depend upon the type of
embankment:
¼In a modified homogenous dam, the filter is generally placed as a blanket15
of sand and fine gravel on the downstream foundation area, extending from
the cutoff/core trench boundary to the edge of the downstream toe and then
taken to safe discharge by the toe drains.
¼In a zoned dam, the filter is placed between the core and the downstream
shell zone. A longitudinal ‘chimney’ drain of gravel material that collects the
intercepted seepage flow and carries it to the base of the chimney and, via
one or more transverse drains, conveys the water to the toe drains outside the
embankment.
Such drains are essential when seepage risks are considered high – for example,
a downstream fill material of fairly low permeability, or a homogeneous dam
on an impervious foundation, would always require seepage drains. A saturated
downstream area can lead to instability and slippage. If this is significant it may
deplete the volume of fill to the extent that the weight is insufficient to resist the
forces exerted on the embankment by the water pressure in the reservoir and from
beneath the dam. Partial or complete failure may then result.
Even if an environmental impact assessment is not required, at the design stage for
any new dam, consider the need for environmental flows and releases from the
dam – usually in the dry season – to maintain the downstream watercourse in as
natural condition as possible.
Provision of drinking water supplies downstream of the dam (using pipes under
or through the embankment and simple, sand filters and stand pipes under gravity
pressure) will reduce access to the reservoir by people and livestock. Alternatively,
wells and hand pumps in the same area may prove suitable and allow local people
access to water that may otherwise be lost to seepage.
Fencing the dam and reservoir may be required to prevent access to the embank-
ment and reservoir. Where this is not possible and to reduce the incidence of shisto-
somiasis, malaria and other water-borne diseases by keeping grass cover around the
reservoir and in flowing channels to a minimum (including regular cutting), raising
and lowering reservoir levels and removing the possibilities of standing water in
and around the dam will help.
6 Detailed design 57
Much of the above should become the responsibility of the communities benefiting
from the dam and a programme of education (incorporating health and sanitation)
on the use of the dam and its resources should be initiated at an early stage in the
design/construction process. Involvement of the beneficiaries in any remedial or
mitigation works (under any community contribution to the overall works) also
engenders a sense of responsibility in using and maintaining the water resource
provided.
Dam construction disturbs the landscape around the dam (excavation, clearing areas
for storage, accommodation and parking, access roads) and such works should be
kept to a minimum. It should be part of any contract for the contractor to remove
and store the topsoil of any area to be disturbed and then return such topsoil to
the site to allow normal vegetation to re-grow and prevent any subsequent erosion.
For borrow areas it can prove difficult to restore them to their original condition
but infilling them with waste material from the dam reservoir area and then topsoil-
ing and grassing them will mitigate much of the negative impacts. Alternatively,
converting any such pits to fish ponds can be considered (and the pits can be exca-
vated at the time of construction with this eventual aim in mind).
Dam construction 7
7 Dam construction 61
7 Dam construction
The centre-line pegs should be installed at the ends of the embankment and at
every change in ground level. For each change in ground level a ‘mating’ peg (see
Figure 11a) should be established by level or GPS on the opposite side of the
valley, but still on the centre line.
At each peg on the centre line of the embankment, the distances of the toe pegs
upstream and downstream are calculated and set out at right angles as in Figure 11b.
The toe peg offset distances from the centre line are calculated using the formula:
Offset distance (m) = S. H + 0.5 Cw
On the spillway side, pegs are located where the spillway cut (if any) begins and
ends and additional pegs are placed in an arc along the sides of the spillway channel
(see Figure 11b). A 15 m interval between pegs is desirable and each should show
the depth of the excavation required, note being made of the slope within the
spillway itself (usually 1:400) needed to encourage flood water to flow away from
the training bank and end of the embankment.
When all the pegs have been installed, and a full pegging layout drawn up, all the
ramifications of the project can be discussed with the client and/or plant operator so
that any risk of error and opportunity for misunderstanding are minimized and use
of equipment and efficiency maximized.
62 Manual on small earth dams