176 The N.Z. Journal of Science and Technology. [Aug. 
be available so near the present one that no correcting factor will be 
required. But in any case arrangements should be made for at least one 
year’s overlapping observations at the old and new sites to establish the 
correcting factor if necessary, and in future every effort should be made 
to select the best site for the meteorological station and to render it 
permanent. 
So far only the results for individual years have been considered. But 
the risks involved are obviously very much increased in the case of 
consecutive years of excess or deficient rainfall. For the agriculturist the 
soil, and for the hvdro-electric engineer the reservoir storage reserve, can 
probably stand one year of drought without serious loss, but a second or 
'30 
n • 
f 
n 
i i 
/20 
/ 
\ 
M 
r 
\ 
,r—\ 
T 
i 
r 1 
i 
i 
i 
M 
r 
i 
i 1 
no 
- 
f 
t 
J 
\ 
\ 
! 1 
i 
i 
1 
1 
i r 
i i 
7 \ 
\ 
i 
i 
1 
1 1 
A 
\ r 
i i 
J i 
r-' 
i 
1 
i 
i 
mo 
TT 
j_L 
\ 
L \ 
\_ 
! 
i 1 
j| 
l 
r-l 
/ 
\ i 
u 
”T“- f 
\ 1 
u 
i 
a 
90 
\ i 
L.l 
L 
J 
i 1 
u 
80 
—SOL 
/TH ISL/9HD 
— 
. 
7° 
60 
JO 
40 / 
30 
r 
-A 
20 
\ 
i 
\ i 
v t 
\ / 
, / 
\ / 
\ 
/ 
\ 
r 
_/ 
r 
s 
“ 
IO 
l /* 
L-./ 
\ 1 
\ i 
\ i 
\ / 
0 
- Hokitika 
- Lincoln 
- Dunedin 
1892 94 36 98 1900 02 04 06 08 19/0 12 t4 /& IS 1920 
Fig. 3. 
third year of deficient rainfall may involve financial disaster. It is there¬ 
fore of even greater importance to know what has been the frequency in 
the past of the occurrence of consecutive years of high and low rainfall. 
Table II herewith gives the average cycle over which excesses and 
deficiencies of 10, 20, and 30 per cent, in rainfall have occurred in the 
biennial and triennial periods for the three stations, Auckland, Wellington, 
and Dunedin. The triennial periods are, of course, much less frequent, 
though it might be noted that Wellington suffered from excess rainfall 
for three consecutive years in 1877-79, and Auckland in 1915-17, and 
Dunedin suffered from deficiency in rainfall for three consecutive years in 
1889-91 and again in 1914-16, as shown in fig. 1. 
The influence of location and geographical position on the rainfall, 
showing the enormous difference in the average rainfall as between, for 
instance, Hokitika (114*13) and Lincoln (25-34), does not come within the 
scope of the present investigations, but is worthy of further consideration 
(fig. 3). 
