Page 49 - Water Engineering Hydraulics, Distribution and Treatment
P. 49
27
2.5 Loss by Evaporation, Seepage, and Silting
Solution 2 (SI System):
3
An SI or metric system solution can be obtained using these conversation factors: 1 MG = 3785 m = 3.785 ML;
2
2
2
3
3
2
3
1ML = 1000 m ;1 MGD = 3785 m /day = 3.785 MLD = 0.0438 m /s = 43.8L/s; 1 mi = 2.59 km ;1 MG/mi = 1.461 ML/km ;
2
2
1 MGD/mi = 1.461 MLD/km .
the difficulties and costs associated with shortage in sup-
reception of rainfall; (b) rising and falling water levels alter
ply. Storage values equaled or exceeded but once in 20, 50,
the pattern of groundwater storage and movement into and
or 100 years, that is, 5%, 2%, and 1% of the years, are
out of the surrounding reservoir banks; (c) at high stages,
often considered. For water supply, Hazen (1956) suggested
water may seep from the reservoir through permeable soils
employing the 5% value in ordinary circumstances. In other
words, design storage should be adequate to compensate for evaporation to the atmosphere and gains water by direct
into neighboring catchment areas and so be lost to the area
a drought of a severity not expected to occur more often than of origin; and (d) quiescence encourages subsidence of set-
once in 20 years. In still drier years, it may be necessary tleable suspended solids and silting of the reservoir.
to curtail the use of water by limiting or prohibiting, for
example, lawn sprinkling and car washing.
Restricting water use is irksome to the public and a poor 2.5.1 Water-Surface Response
way to run a public utility. As a practical matter, moreover, The response of the new water surface is to establish new
use must be cut down well in advance of anticipated exhaus- hydrologic equilibria (a) through loss of the runoff once
tion of the supply. It would seem logical to consider not only coming from precipitation on the land area flooded by the
the frequency of curtailment but also the depletion point at reservoir Qa (closely), where Q is the areal rate of runoff of
which conservation should begin. In practice, the iron ration the original watershed and a is the water surface area of the
generally lies between 20% and 50% of the total water stored. reservoir; and through evaporation from the water surface
Requiring a 25% reserve for the drought that occurs about Ea, where E is the areal rate of evaporation; and (b) through
once in 20 years is reasonable. An alternative is a storage gain of rainfall on the water surface Ra, where R is the areal
allowance for the drought to be expected once in 100 years. rate of rainfall. The net rate of loss or gain is [R − (Q + E)]a;
This is slightly less in magnitude than the combination of a a negative value records a net loss and a positive value a net
25% reserve with a once-in-20-years risk.
gain.
In undeveloped areas, few records are even as long as
Individual factors vary within the annual hydrologic
20 years. Thus, estimation of the 5%, 2%, and 1% frequen-
cycle and from year to year. They can be measured. Exact
cies, or of recurrence intervals of 20, 50, and 100 years,
calculations, however, are commonly handicapped by inad-
requires extrapolation from available data. Probability plots equate data on evaporation. Required hydrological informa-
lend themselves well to this purpose. However, they must be tion should come from local or nearby observation stations,
used with discretion. Where severe droughts in the record areas of water surface being determined from contour maps
extend over several years and require annual rather than of the reservoir site. The mean annual water surface as a
seasonal storage values to be used, the resulting series of fraction of the reservoir area at the spillway, f, is normally
storage values becomes nonhomogeneous and is no longer about 0.90 or 90%.
strictly subject to ordinary statistical interpretations. They For convenience, the water-surface response is expressed
can be made reasonably homogeneous by including, besides in one of the following ways:
all truly seasonal storage values, not only all true annual stor-
age values, but also any seasonal storage values that would 1. Revised runoff Q = Q − (Q + E − R)(fa/A) (2.2)
r
have been identified within the periods of annual storage
2. Equivalent draft D = (Q + E − R)(fa/A) (2.3)
if the drought of the preceding year or years had not been e
measured. Plots of recurrence intervals should include minor 3. Effective catchment area A = A − fa[1 − (R − E)/Q]
e
storage capacities as well as major ones. The results of these
(2.4)
statistical analyses are then conveniently reduced to a set of
draft–storage–frequency curves. 4. Adjusted flow line F = Q + E − R (2.5)
5. Effective draft D ed = D md + D (A) (2.6)
e
2.5 LOSS BY EVAPORATION, SEEPAGE,
where Q = revised runoff, in./year or cm/year; Q =
AND SILTING r
mean annual runoff, in./year or cm/year; R = mean annual
When an impounding reservoir is filled, the hydrology of the rainfall, in./year or cm/year; E = mean annual evaporation,
2
inundated area and its immediate surroundings is changed in./year or cm/year; a = reservoir area, mi 2 or km ;
2
2
in a number of respects: (a) the reservoir loses water by A = catchment area, mi or km ; f = 90%= 0.9 = effective