Diurnal Weather Patterns— Leopold 
85 
NUMBER OF RAINFALL OCCURRENCES AT VARIOUS HOURS 
FIG. 3 A KAWAIHAPAI 
INDIVIDUAL HOURS JAN. 1945 - JAN. 1947 
J1 
__ INDIVIDUAL HOURS JAN. 1945 - JAN. 1947 Jj. 
AAAjj "Aj A 
>H SIGNIFICANCE LEVEL 
1 r 
FIG. 3B KAWAIHAPAI ' 
8 HOUR SUMS JAN. 1945 - JAN. 1947 
~ 1 
LEVEL 
.. 
I 
24 0 
12 
24 0 
12 
24 0 
~~t —i f 
FIG. 3C LEILEHUA 
MAY I, 1943- 15 
r 
12 
60 
40 
20 
24 
HIGH SIGNIFICANCE 1% LEVEL 
SIGNIFICANC E^LEVEL 
.A 
80 
60 
40 
20 
150 
100 
O 12 24 0 12 24 0 12 24 
DEC.-JAM-FEB. MAR.-APR.-MAY JUNE-JULY-AUG. 
HOUR OF DAY 
Fig. 3. Number of rainfall occurrences at various hours at three stations on Oahu. 
The maximum number of occurrences which 
might, by pure chance, fall into a time period 
of 8 consecutive hours is plotted as a long- 
dashed line in Figure 3B at the ordinate value 
computed. The minimum number which might 
fall in an 8-hour group by chance is similarly 
represented at the computed lower ordinate. In 
the same way the number of occurrences ex¬ 
pected once in 20 trials is shown by short- 
dashed lines at appropriate ordinate values. 
These computed numbers required for signifi¬ 
cance (adjusted X 2 = 3.841) and high signifi¬ 
cance (adjusted X 2 = 6.635) obviously depend 
in part on the total number of rainfall occur- 
