Waikato Control Plan Studied

Press, Volume CVI, Issue 31081, 9 June 1966, Page 6

Waikato Control Plan Studied

(from the London correspondent of “The Press”) LONDON. Work on a model of the lower Waikato river, which began at the Ministry of Technology’s Hydraulics Research Station, Wallingford (Berkshire), early last year, has now reached a critical “proving” stage.

The model is completed, and is being adjusted to reproduce conditions recordded in the past on the Waikato. When the scientists working on this £21,800 project are satisfied with the model’s behaviour, they will begin testing the effects of the Waikato Valley Authority’s flood control plans. It is expected that the main testing will begin next month and be completed by the end of the year. Progress results are being sent to New Zealand and a draft report may be ready next March with the final result of the experiment available about June next year.

The model occupies an area roughly 160 ft by 50ft, but would have taken up more of the research station’s vast experimental hall if the geographical run of the river had not been “bent” back upon itself in the middle (roughly near Mercer) and a pool to represent the action of Lake Waikare fitted like a piece of jigsaw puzzle in an empty space between Port Waikato and the river mouth on the one side and the beginning of the model at Ohinewai, four miles downstream from Huntly, on the other. Such liberties with nature do not affect the current experiments. But the general course of the river, its width and depth have been most accurately moulded to exact scales— Ito 720 horizontally, 1 and 1 to 100 vertically.

The horizontal reduction was made for a suitable size of model at the research station. The vertical reduction is not so great, because depths of water running in the model would become so small that the flow would be unduly slowed by friction. First Stage

The first stage in the making of the model after deciding on the scale and working out the design on paper, was the construction of the tank and floor to contain it. In November last year, working from maps, levels and aerial photographs provided by the Waikato Valley Authority, the Station began to make wooden templates corresponding to the cross section of the river bed at various places. These were accurately aligned on the floor of the model and rubble and sand packed between them with a top layer of two inches of concrete up to the level of river bed.

Off-chanel areas where a flow goes on in high flooding were then constructed working from tables of ponding volumes rather than geographical outlines.

The river-bed is then roughened by insertion of a large number of small pebbles. Small metal plates drilled with fine holes are also added in some places to represent the hindrance to flow offered by willow trees. During the proving pebbles and plates may be added or taken away in places until the flow is properly adjusted. The first proving tests have been comparing the model’s water depths with the March, 1964, and July, 1961, Waikato floods, putting in representative volumes of water. Now the scientists are beginning the main proving based on the design flood recordings of 1958. Flow Factors At present there are 20 elaborate electronic water depth recorders in strategic places over the flow of the model. Each is linked to a central data processing unit. At the touch of a button the researchers can obtain a printed tabulation of the depths at each part of the river at a specific time. Depths of the river vary in real conditions according to rainfall and other factors affecting the flow along its length and the tidal effects at the river mouth affecting the outflow. These effects, too, are carefully reproduced in the model

by means of special hydrographs which regulate the amount of water pouring, into the head of the model and reproduce tides by means of raising and lowering a weir outside the river mouth. A 24-hour pattern of tides occurs in 20 minutes in the model, and the hydrographs go through a full 15-day cycle in five hours.

Timing becomes all important once an experimental run begins and the researchers use stop-watches to obtain depth results from the data processing unit at correct intervals. Once the station is certain the model is behaving in the same way as the Waikato did in the various known floods they have reproduced, they will go on to duplicate the new stop-banks planned by the valley authority and the removal of various willow trees and other obstacles in certain places. Realignment of the Great South road near Rangiriri, an important feature of flood control plans, will then be duplicated in the model, and probably several' heights of roadway will be tested. At 55,000 cusecs it is planned that water will flow over the road for about a mile, but only at a depth of some six inches so that traffic will not be stopped. Floodgates will stop the flow of the Rangiriri stream during floods and the Te Onetea stream is to be blocked off completely. The water will then flow back to the ponding lakes. The effect of all these different schemes will be experimentally determined, (using the 1958 flood volumes to test them) and at the end of the experiment, the Station will investigate the effect of dredging channels in the tidal reaches of the river and possibly cutting a new channel at the river mouth. (The scientists say that the expense of this measure would be so great that results would have to be dramatic to justify it.) A “mobile bed model” of the river will then have to be constructed if the authority decides to proceed (at extra expense) to investigate the effects dredging would have on silting in various places. A discharge of one cusec (one cubic foot a second) of water in the model, would represent 720,000 cusecs in the Waikato. In testing the 1958 flood conditions (the highest flood since 1907) the scientists are only dealing with flows up

to 65,000 cusecs so that onetwelfth of a cusec flows in the model.

Nevertheless the water-depth recorders read to an accuracy of one-thousandth of a foot in the model, which represents one-tenth of a foot in the actual river, so that the allimportant water-levels of the Waikato should be determined quite accurately to an inch. Most flow at the peak of a flood is in the Waipa (about 40,000 cusecs) and only about 27,000 cusecs comes from the

Upper Waikato (largely controlled by hydro-electric schemes). The design flood is reached by a combination of these two flows and allowing a lag time corresponding to the arrival of storms in different places in 1958 60,000 cusecs at Ngaruawahia —plus a further 5000 cusec allowance for a peak flow from the Mangawara at Taupiri. This makes 65,000 cusecs flowing down through Huntly, and into the area covered by the model.