за
36
HYDROLOGY.
The problem of water supply is inseparable from the study of the hydrology of the district. Unfortunately there are not many records available in Hong Kong, and the few collected before the war have now been lost with the result that it is frequently necessary to work backwards to reconstruct the design figures for a given scheme.
Hong Kong having marked wet and dry seasons, for the purposes of water supply it is preferable to consider the rainfall for one com- plete wet and dry cycle rather than the calendar year. Selected years were therefore chosen from the figures published by the Royal Obser- vatory and re-written in the order May to April, with some interesting results:--
(a) The maximum rainfall for this twelve monthly period was 128.37 ins. in 1891/2 instead of 119.71 ins. for the calendar year 1889.
(b) The driest year 1895/6 produced 53.81 ins. as compared with
45.83 ins. in 1895.
(c) The average of the 4 wettest years is 111.37 ins. against 103.07
ins.
(d) The average of the 4 driest years is 63.64 ins. against 64.8 ins.
These results are all in favour of the waterworks, and the most important fact is that the average for 4 dry years is not substantially altered.
Hong Kong's waterworks therefore should be designed to tide over four dry years i.e. four successive years when the rain is less than average, during which the average annual rainfall is about 64 ins. The difficulty, however, is to assess the yield from the rainfall. This is further complicated by the uncertainty as to the value of the indirect catchment areas, that is areas not draining naturally into the reservoirs but from which some proportion of the run-off is diverted to the reservoirs by catchwaters.
These figures are unfortunately not at present determinable due to the lack of recording flumes on the catchwaters, but overall figures are obtainable from the recorded storage position. Taking a fairly
dry year 1936/7 when the rainfall was only 69.56 ins. the yield from the 6030 acres on the Island was 3660 M.G. or approximately 40%. This problem will be referred to later under the heading of catchwaters.
Apart from the yield to be expected, the other important factor to be determined is the maximum flood discharge for which the dams must be designed. There is considerable uncertainty as to the value
į
I
37
of this, particularly as a downpour of 5 ins. in one hour is known to have occurred in the Colony, but there is no information as to the run-off experienced from that storm. Nor can this be very accurately estimated from predictions of rainfall and run-offs, as the few records available show large variations in intensity over very small areas. The only satisfactory solution therefore is to observe the rainfall from as many raingauges as possible, including at least one continuous recording gauge, together with the actual run-off as recorded by stream gaugings.
The author is at present engaged on a study of the Tai Lam Chung stream in connection with the proposed new reservoir to be sited in that valley. A 60 ft. full flow weir has been constructed with provision for recording heads up to 10 ft. above the cill. Unfortun- ately the recording apparatus was not ready this year so that no accurate flood figures are yet available. A further complication arose from the fact that being in a stream course there is only a compara- tively small reservoir above the dam with the result that in flood there is a considerable velocity of approach which renders all available discharge formulae inaccurate. To calculate the discharge therefore, two small weirs have been set up in a branch stream, one an ordinary V-gauge whose discharge characteristics are known and below it is a 1:10 scale model of the 60 ft. weir.
A certain amount of information however has been gathered about the dry weather run-off. During last winter the minimum run-off was as low as 200,000 gallons per 1,000 acres per day, and in the seven months October to April inclusive the average was only 780,000 gallons, which allowing for evaporation and other losses which would occur in a reservoir, would probably mean a yield of only 0.6 M.G.D. per 1,000 acres of catchment.
As a check on this figure the Island reservoir records were analysed for the same period with a rather surprising agreement, the corresponding figure being 0.56 M.G.D. The Kowloon Reservoirs and Shing Mun jointly gave a figure of 0.84 M.G.D. a little higher, which was rather to be expected.
These figures agree remarkably well and will therefore be used later in a discussion on the available resources, but it should be noted that they are not taken over a long enough period to form permanent guides, and may have to be amended after further study.
RESERVOIRS.
At the present time therefore, the reservoirs of the Colony consist of five separate, though to a certain extent interconnected, systems. They are:-
40