296 The N.Z. Journal of Science and Technology. [Sept. 
belief that the ratio was 100 per cent, of the rainfall. Experience from 
accurate gaugings of the Wairua River also bear out the possibility of such 
a ratio. In Table III is shown the ratio of run-off to rainfall for the Wairua 
River for each month of 1915. 
Table III. —Percentage of Run-off from Wairua River for the Year 1915. 
(Catchment area, 278 square miles.)* 
1915. 
1 
Jan. Feb. 
1 
Mar. April 
May. 
June. July. Aug. 
Sept. Oct. 
Nov. 
Dec. 
Rainfall, in inches 
1-50 0-57 
0-58 
3-39 
4-41 
4-23 7-00 8-54 
7-22 6-16 
315 
0-82 
Run-off, percentage 
14 42 
16 
26 
62 
81 104 .. 
68 102 
71 
76 
* Rainfall is average of three rain-gauges. August record of river-heights destroyed. Run off, 
August, 1914, 98 per cent. Total rainfall, 1915, 52 - 5 in. Average run-off for year, 63 per cent. 
It will be seen from this table that in the wet months of the year the 
ratio of run-off to rainfall is over 100 per cent. The yield of the river for 
each month was calculated with considerable accuracy from charts showing 
automatically the height of the river, and by means of a graph giving the 
discharges corresponding to the heights. Unfortunately, the quantity of 
rainfall is not so accurately determined, as it is the arithmetical mean 
of only three rain-gauges, which, although well distributed, cannot give a 
close record of average rainfall over an area of 278 square miles. The 
figures give, however, a fair indication that the run-off in wet seasons 
must be nearly equal to the rainfall. The condition of the ground in 
February, 1917, would be very similar to that of July to October in any 
other year. The average rainfall in the Mangakahia Valley for eight 
months before January, 1917, was 10-4 in. per month. 
With regard to a ratio of run-off exceeding 100 per cent, of the rainfall, 
Mr. W. C. Reid, in a paper read before the Institution of Civil Engineers,* 
gives from a catchment area a yield per month frequently in excess of the 
amount of rain precipitated. 
Considering the run-off from the Mangakahia River, Table II shows a 
run-off equal to 1-13 in. per hour. Table I shows that 10 in. of rain fell 
in two days in this valley ( i.e ., at Keretoki and Waimatenui). The rain 
ceased at 1 p.m. on the 3rd February, and, as the records are taken at 9 a.m. 
each day, 10 in. of rain, fell in twenty-eight hours. This fall is nearly equal 
to 9-09 in. in twenty-four hours at Taheke on the 2nd February. As rain 
ceased at Keretoki at 1 p.m. on the 3rd and we have a record of 4*57 in. 
on that date, this rain must have fallen at the average rate of 1-14 in. per 
hour for four hours. The station observer at Waimatenui states that he 
recorded 6 in. of rain in three hours, between 8 a.m. and 11 a.m., on the 
3rd February, a rate of 2 in. per hour. If, as I assumed above, the run-off 
is equal to the rainfall, the records obtained of the flood-discharges seem 
to agree fairly well with the rainfalls recorded. 
It is interesting to note that the curve showing rainstorm-intensities 
for varying periods of rainfall at Sydney, N.S.W., gives 2 in. per hour for 
a storm of three hours duration, and 1 in. per hour for a storm up to eight 
hours duration. A similar curve for Melbourne, Victoria, gives 0-66 in. 
per hour for a three-hour storm and 0-5 in. for a four-hour storm. 
* Trans. Inst. Civ. Eng., vol. 194, May, 1914, pp. 3-19. 
