Nov. 8, 1915 
Use of Current Meters in Irrigation Canals 
219 
with integration methods are shown, as measurements by this method 
were not taken in all cases. 
Table I .—Variation in discharge in percentage by the two-point , the single-point, and 
the integration method, compared with the multiple-point method 
Two-point method. 
Single-point method. 
Integration method. 
Type of canal cross section. 
Num¬ 
ber 
of ob¬ 
serva¬ 
tions. 
Mean 
differ¬ 
ence 
from 
mul¬ 
tiple- 
point. 
Aver¬ 
age 
varia¬ 
tion 
of a 
single 
obser¬ 
vation. 
Num¬ 
ber 
of ob¬ 
serva¬ 
tions. 
Mean 
differ¬ 
ence 
from 
mul¬ 
tiple- 
point. 
Aver¬ 
age 
varia¬ 
tion 
of a 
single 
obser¬ 
vation. 
5 Per 
cent 
correc¬ 
tion ap¬ 
plied. 
Num¬ 
ber 
of ob¬ 
serva¬ 
tions. 
Mean 
differ¬ 
ence 
from 
mul¬ 
tiple- 
point. 
Aver¬ 
age 
varia¬ 
tion 
of a 
single 
obser¬ 
vation. 
Rectangular flumes .... 
27 
+0.68 
1.45 
27 
+4.90 
2. 21 
*7 
+ 1. 06 
I. 36 
Concrete-lined trape- 
zoidal sections. 
15 
+ .86 
I. 42 
15 
+ 4. 21 
I. 94 
4 
+ • 72 
•93 
Shallow earth canals, 
sloping sides. 
13 
- .38 
I. 08 
13 
+ 3 - n 
3 - 42 
9 
- .81 
2-44 
Shallow earth canals, 
steep sides. 
25 
+i- 05 
i- 74 
25 
+ 5. 02 
2. 44 
18 
+ .36 
2. 15 
Earth canals, relatively 
deep sections . 
1 6 
+ 1. 07 
1. 70 
15 
+6. 32 
3. is 
7 
+3. 06 
3 - 78 
Mean of all. 
96 
+ • 73 
*• 5 1 
95 
+4. 80 
2. 54 
55 
+ . 76 
2. 07 
Table I shows all three methods to give an average discharge greater 
than the multiple-point gaging. For the two-point and integration 
methods this is not large, being about three-fourths of 1 per cent for both 
of these methods. For the single-point method the average error is 
+ 4.80 per cent. This is large enough to warrant a correction factor, so 
that all further comparisons with this method are based on a correction 
of — 5 per cent made to the discharge secured by the single-point method. 
Besides the average error of the series of experiments, the probable 
or average variation of a single observation is also given. While the 
mean difference of the two-point and integration from the multiple-point 
method is the same, the single measurements show a somewhat greater 
average variation for the integration than for the two-point method. 
If the results of the single-point observations are reduced by 5 per cent, 
the corrected results have an average variation but little in excess of the 
other methods. These results may be expressed by saying that with the 
two-point method a series of observations will give results three-fourths 
of 1 per cent too high. If no correction is made to the results, single 
observations will have an average error of 1.5 per cent. 
The experiments covered a wide range of discharges and canal types, 
so that further classifications were made to determine the effect, if any, 
of differences in the velocity, the depth, or the value of n on the accuracy 
of the different methods. The results are given in Table II. 
