206 
A NEW METHOD OF ESTIMATING STREAM-FLOW 
on Stream A and 0.029 cubic foot per second on Stream JB (28 and 30 per cent of 
the constant part of the flow, respectively) is classed as a flood month. On this 
basis it appears that there are 26 flood months among the 58 on Stream A and 24 
on Stream B. The agreement between the computed and observed daily flow in 
non-flood and flood months is shown in Table 55. 
It is clear that the flood-flows are much less accurately computed than the 
normal flows. 
COMPARISON OF 1914-15 WITH 1911-12-13 
Is the agreement between computed and observed stream-flow much poorer 
during the years 1914 and 1915, which were not considered in the computations, 
than during the years 1911-12-13, from which the constants were computed? This 
agreement is shown in Table 56. 
Table 56 — Comparison of daily computed and observed flows in 1911-12-13 against 1914-15 
Percentage of days on which computed flow agrees with 
observed flow 
Within 10 per cent 
Within 20 per cent 
Within 50 per cent 
Stream A 
1911-12-13 
1914-15 
37 
17 
29 
48 
64 
40 
63 
68 
89 
81 
94 
93 
1911-12-13 
1914-15 
The comparison on Stream A is against the period 1914-15. Some of the 
comparisons to be given later show closer agreement in 1914-15 on both streams 
than in the earlier period from which the constants were computed. 
COMPARISON BY MONTHS 
For many purposes it is more important to compute the mean flow accurately 
for a month than to compute it accurately for each day. A comparison has, there- 
fore, been made between the computed and the observed values of the mean stream- 
flow in each month. This is shown in Table 57. 
Table 57 — Comparison of computed and observed flows by 7nonths 
Percentage of months in which computed flow agrees with 
observed 
Within 10 per cent 
Within 20 per cent 
Within 50 per cent 
31 
28 
62 
62 
88 
97 
Stream B 
SPRING FLOOD COMPARISONS 
Even a casual examination shows that the larger monthly discrepancies usually 
occur at the spring flood-flows. This naturally leads one to a special study of the 
spring floods. The displacement in time between the computed and observed 
