WMOD00519

• Project Name: Weather Modification Programme - Precipitation Enhancement Project
• Title: A Review of the Hydrological Aspects of Evaluation of Precipitation Enhancement
• Date: 5/1/1977
• Weather Modification - Doc Type: Report

2

6. As an alternative to this costly process, designers of evaluatj,on systems
for precipitation enhancement projects have considered using comparisons of treated
and untreated streamflo~1 response to precipitation, In regions where increased
"later supply is frequently desired, car'efully made streamflo", observation:3 may be a
long-established part of the 'vater management system. rJ~he streamflo", past a g'auging
point is an integrated resultant of the varying conditions over the contributing
drainage area, As such it provides, in a single number, a means of characterizing
the precipitation experience over what may be a largely inaccessible area without
requiring that the point-by-point details of that precipitation experience be kno"m.
In addition, as the work of Crawford and Linsley (1963) has shown, a given increase
in precipitation may produce a greater percentage increase in streamflo~, thus making
detection of the increase easier.

I
i
f
I
I ,
I I
I

investment required to provide the large number of gauges required, the job of
collecting, processing, and analysing the mass of precipitation data generated by a
dense gauging network requires commitments of major amour,ts of timE and money.

7. There are, however, disadvantages to the use of streamflow for evaluation
that may ouhleigh the adve.ntages in some circumstances. Inaccuracy of discharge
measurement, high variability of natural flo"IS, and time dependence of successive
river flo",s have been identified (Thorn, 1957; Harkovic, 1966) as the min disadvant-
ages. These win be discussed briefly as an aid to those decision-makers in PEP
and elsewhere "Iho may not have been previously concerned vri th hydroloBY'

i I

III BACKGR01JND INFOHHATION ON THE NATURE OF STRE1\.MFLOW AND ITS 11EASUREMENT

8. The accuracy of streamflow measurements has been addressed (DiCkinson, 1967)
as part of a comprehensive project to develop techniques for evaluation of precipita-
tion enl1ancement projects. His work provides an objective means of evaluating the
import of streamflow errors.

9. Streamflo\v volume is computed by integrating a time series of discharge
rates. Discharge is usually measured by observing the depth of water flowing past a
known cross-section. This cross-section may range from the calibrated notch in a
weir, as used in the measurement of small flows, to the entire cross-section of the
channel of a major river. Heasurements of depth are converted. to discharge values
by entry into a rating curve "Ihich relates discharge to water depth. l!'or the notched
",eirs and flumes used to measure discharge in small streams and irrigation canals,
the rating curve can be quite precisely determined and remains reasonably stable.

10. For larger streams, the rating curve is the result of sampling the flow
rate vrith a current meter at a number of vertical profiles across the width of the
channel. Since the flow rate varies in the vertie-al profile due to friction effects
and turbulence, sampling errors are possible. The number of profiles taken across
the width of the channel also affects the sampling error. In the time required to
observe a number of profiles, the stage or depth of the river may change. The dis-
charge at a given stage will differ, clepending on "lhether the river is rising or
falling. The final accuracy of a ratine curve will be influenced by the condition
of the floH meter, the turbulence of flo"l, the sampling procedure, ho\v \'Iell the
assumption of steady-state flow during the transect is met, the magnitude of the
difference in rising and falling flow (Fread, 1976). In addition, rating curve accu-
racy depends on how stable the cross-section formed by the riverbanks and bottom
remains during the rating process and for subsequent use of the rating curve. If the
channel is formed of sandy banks and bottom, major changes in c:J:'oss-section can occur
during the course of a rise, fo110",ed by a return to original conditions after the
flow subsides. A rocky channel usually offers acceptable stability, at least unb.l
flo\'ls become energetic enough to move the rocks.