Art. VI.—On the Taieri River Floods and their Prevention.
[Read before the Otago Institute, 22nd November, 1881.]
Immunity from destructive floods during the last few years, together with the extension of the Taieri River embankments, appear to have induced a sense of security amongst the settlers on the Lower Taieri Plain, which may at any time be somewhat roughly disturbed. And I think, therefore, that the conservators of the river should not remain satisfied with what they have already done, but should take steps to determine accurately and
finally the practicability of utilizing such natural sites as exist for the storage of flood waters, but which have not yet been surveyed and reported on. A tolerably intimate knowledge of the basin of the Taieri River above Outram, and of that of the Waipori River, has satisfied me that certain flats on the banks of these rivers offer very excellent facilities for the construction of dams capable of impounding the most of the flood waters, and that at much less cost than is usually supposed. The expenditure of about £200 on a survey would settle definitely this question, and the money would be well spent.
Floods on the Taieri River have been reported on, or written about, by Mr. J. T. Thomson, C.E., Mr. G. M. Barr, C.E., Messrs. Blair, Bell, and Higginson, M.I.C.E. (Flood Commissioners), and Mr. E. B. Cargill. Their estimated discharges of the river and its tributaries differ somewhat, as well as the remedies propounded; and as to the scheme of the Flood Commissioners, its cost has been fatal, apparently, to its early realization. I propose, therefore, to give you a few facts in support of my statement that the Taieri floods are capable of being checked, if not entirely reduced to manageable limits, by reservoirs, and that at a very moderate expenditure.
The opinion that the Taieri floods are due to the sudden melting of snow, caused by warm north-west winds, with or without rain, is erroneous. For, first, in the case of the gathering ground of the snow-fed rivers of Otago—the Waitaki, Clutha, Oreti, and Waiau rivers—the greater part of the snow lies above the 4000 feet level, while the mountains which discharge the greatest amount of water into the Taieri—viz., the Rough Ridge, Lammerlaw, and Rock and Pillar Ranges—do not exceed 3800 feet in altitude (in round numbers), with the exception of about 3000 acres on the Rock and Pillar. The snow is generally off the latter mountains by the middle of November, and this season, while I write (October), there is little or none; and the Taieri floods, particularly the most destructive, do not generally occur while snow is on the highest catchment ground, that is, in winter. The worst flood of all, that of 1868, occurred in the beginning of February. Lastly, the heaviest floods have been consequent on rain-storms from east to south-west. This range of direction does not bring warm rain; that comes from north-west; but the storms blow well home, and as they reach the Upper Taieri Plains keep pretty steady at south-east. My own notes on the weather, made while living on these plains during 1867–68, and 1869, show this to be correct. It is corroborated by the experience of settlers living there, and the meteorological observations taken at Dunedin for the same period are confirmatory. That is the first circumstance to be kept in view in ascertaining the distribution and effects of the rainfall.
Next, the configuration of the catchment area in its main features is of great importance. This consists of the Mount Ida range and the Kakanuis on the north; the Rock and Pillar, Lammerlaw, and Rough Ridge on east, south, and west; and the Upper Taieri or Maniototo Plains in the centre, which latter average about 1400 feet above sea-level. And, lastly, the soil generally over the whole area is dry, in many places shingly, and, as a whole, has become so baked on the surface by grass fires and stocking that the rainfalls run off with great rapidity—greater, it is thought by many, than was the case before the interior was settled. But there is one remarkable exception to this general character of the soil. The plateau, or elevated plain which forms the top of the Lammerlaw and Rock and Pillar Mountains, contains a large morass of 3000 acres in extent, besides numerous smaller swamps, with bogs and lagoons. This sponge like and porous soil, with lagoons, is more or less characteristic of the whole catchment area of the Taieri, above the Styx Stream, including the Serpentine Flat, 1800 feet above sea-level.
Bearing in mind, then, these facts—the direction of the rainstorms which, at their worst, come from S.E. to S.W., the exposure and resistance offered by the faces of the mountains to these storms, and the nature of the soil on the mountain tops and on the plains, with the relative heights of these localities—what should we expect? We should expect the rain-clouds, which come up in a storm from the Southern Ocean, on reaching our coast, and losing the contents of their lower strata among the coast hills, would pass on until caught by the higher ranges which surround the sources of the Taieri River. There, rolling up the slopes of the Lammerlaw and Rock and Pillar Mountains, and forming an eddy on the flat summits, they would become piled up, and their rate of travelling or velocity being thus reduced, they would naturally deposit the greater portion of their contents on and around the tops of these ranges. The remainder of the rain-clouds would pass on and become gradually dispersed by the higher temperature of the interior plains. The more northerly columns of the rain-clouds would draw along the Horse Range to the highest peaks of the Kakanui Mountains and Kyeburn Hill, and lose the greater amount of their contents among these peaks. At the same time the Maniototo Plain itself, lying immediately west of these ranges, together with its western boundaries, the Mount Ida Range and northern part of the Rough Ridge, would have but a reduced balance of rainfall to receive, reduced still further by the superior warmth of the plain itself.
Evidences of Distribution of Rainfall.
And what do we find to be the case? The experience of the oldest settlers on the Upper Taieri Plains goes to show that the above theory is correct. For instance, Mr. James Murison, who first took up country there as a runholder in 1857, has assured me that, while the southerly storms rage around the Lammerlaw and Rock and Pillar, the basin of the Upper Taieri River itself escapes these to a great extent. From the Kyeburn, all round the north and west side of the plain to the Totara, that is over an area of 280 square miles, or 180,000 acres, there is not a single stream but such as a man may easily jump across. The Kyeburn stream has a rapid descent, and in floods rises to a considerable height, but falls quickly. This I saw during the big flood of 1868, when camped on its banks. The streams then round to the Totara discharge very little rainfall into the Taieri; and the most received by Mount Ida flows into the Waitaki; while on the east side, round to Hyde, there is but a small quantity runs down the Sowburn and Pigburn.* But the Deep Stream and Lee Stream rise rapidly and carry off as swiftly a large amount of rainfall. After the flood of February, 1877, I examined parts of the gorge of the Lee Stream, where the flood marks were visible 40 feet above the ordinary water-level on a width of about two to three chains. This gorge has a descent of 900 feet in 11½ miles, while that of the Deep Stream falls 825 feet in 20 miles, or thereby. Then it is well known the main body of the Taieri above the Styx comes away slowly—owing partly, no doubt, to the sponge-like and retentive nature of its catchment ground, and continues high long after the Kyeburn, Deep Stream, and Lee Stream have run off their flood waters. These latter streams are sudden and violent in their action, especially the Deep Stream, which should be checked; but most danger appears to me to lie in the accumulation of rainfall at the sources of the Taieri itself, after the ground there has become saturated and the river has risen to its full capacity. Here, then, the main reservoir should be.
[Footnote] * I find from measurements recently made by Mr. D. Barron, that the average discharge at Hamilton Bridge exceeds that at Pateroa Ford by only 8,000,000 cubic feet daily.
Stated shortly, the above remarks come to this,—That an excess above the average rainfall on the basin of the Taieri River takes place on and around the Lammerlaw Mountains; that the Sutton, Deep Stream, Lee Stream, and Waipori River, carry off the first of this rainfall, while the Upper Taieri River itself brings away the main flood comparatively slowly after-wards; and that on these streams the necessary sites for impounding reservoirs must be looked for.
Sites for Reservoirs, the Styx Dam.
From a consideration of the above conditions, it seems to me that the drainage of the plain from the Kyeburn round to near the Styx may be disregarded, and that a reservoir at the Styx would catch nearly as much flood
water as one at the Taieri Lake. It is at this place, then, the Styx, that the natural advantages which exist should be utilized by the forming of a main reservoir. Here a large plain, 1800 feet above sea-level, and nearly flat, is narrowed in to a few chains width by the surrounding ridges, which thus offer a natural dam, only requiring man's work to complete it. There is a choice of two sites one mile apart. For nine miles up this plain the total fall on the river does not exceed 40 feet, while the first six miles of this distance, I estimate, has not a greater fall than 10 feet. There is a great width on this plain, which naturally divides itself into two parts, 920 acres in the lower, and 5,344 acres in the upper, or 6,272 acres of water-space for the reservoir all over. The average depth of water, deduced by me from a few levels taken during the triangulation of the Serpentine Flat, are, for the lower area, 27.5 feet, and for the upper area 14 feet, taking the height of the embankment, or dam, at 30 or 33 feet. The capacity of this reservoir will thus be 4,370,636,160 cubic feet, assuming the above figures as correct; and the cost of the dam, 5 chains long, if of masonry, £6390. Now, from the Flood Commissioners' report, I find they calculate on having to impound above the township of Outram, 1,506,400,000 cubic feet daily, while Mr. G. M. Barr's paper gives 4,608,000,000 cubic feet; and a rainfall of 1.5 inches run off over 1700 square miles gives 4,233,968,640 cubic feet daily. The mean of these quantities, or 3,449,456,213 cubic feet, I take as the amount required to be impounded above Outram in one day, and as the duration of flood is found to be from seventeen and a half hours to three days the proposed reservoir has the necessary capacity. For, as the drainage area above the Taieri Lake, compared with that above Outram, is as 850 square miles to 1700, the quantity of water to be retained at the Styx, taking it as the same with that at Taieri Lake, will only amount to 1,725,000,000 cubic feet. So that the Styx reservoir will be far more than equal to the work of retaining this quantity—or it will hold two and a half days' accumulation at the rate of 1725 millions of cubic feet per diem.
The only doubtful element in the above is, does the catchment area above the proposed Styx reservoir run off daily this quantity—1,725 millions of cubic feet of water during flood? This can be ascertained approximately by experiment, and, should it be somewhat less, still the quantity would be large enough when retained by the reservoir, to afford an immense relief to the Lower Taieri Plain.
Deep Stream Reservoir.
Supplementary to that at the Styx, a reservoir or reservoirs on the Deep Stream would be very useful. A good site exists for a dam to one of these, about a mile and a half below Walsh's accommodation house, where an
embankment, 6 chains in length and 15 feet high, would retain 217,800,000 cubic feet over an area of 500 acres. The catchment area above this point is 59 square miles, which, with a rainfall run-off of 1.5 inches, gives a daily discharge of 205,603,200 cubic feet. The cost of this dam, if of masonry, I estimate at £3,115. There are other good sites on this river, but none on the Kyeburn, Sutton, or Lee Stream.
The valley of the Waipori also offers facilities for impounding flood waters above the junction of the Verterburn, and near the outfall of the sludge-channel. I am not so familiar with this spot as with those sites above described, nor do I think it so good. Still, here also there is an area which I estimate at 900 acres suitable for a reservoir. An embankment or dam, 10 chains in length and 33 feet high, would impound about 588,060,000 cubic feet of water, the inclination of the flat being 7 feet per mile. If of masonry, its cost would be £12,000; but probably a careful survey would show that the dam need not be so long, and, consequently, not so costly. This Waipori reservoir would not, of course, be a check on the floods of the Taieri itself, and may therefore be left out of consideration until the money required for its construction is available. At the same time, it would benefit the Henley estate and others of the low-lying lands in that quarter, while indirectly it would relieve the outflow of the Taieri by the lower gorge into the sea.
In each of the above dams, self-acting sluices, openings or culverts, would be necessary, the particular form being a detail which need not here be gone into.
Neither have I particularly referred to the embankment of the Taieri River on the Lower Taieri Plain, from Outram seawards. I would only remark that, while I think it probable the money spent on this might have been expended more profitably in the construction of dams further up the river in the interior, there can be no question as to the benefit the embankment has been to the lands on the right or west bank of the river. This, however, is secured at the expense of the lands on the east side of the river, of the channel of the river, and of the bridges on the river.
In concluding this paper, I would say that the facts I have given go to prove that nearly all the flood-water which could be impounded at the Taieri Lake, may be confined above the Styx River at much less cost, and that an exhaustive survey should be initiated without loss of time, first to determine where the greatest rainfall occurs on the catchment area, and next the precise capabilities of the various reservoir sites which I have indicated, my own calculations of the latter being only approximate,
I append a table of Taieri River floods, data from Dunedin Meteorological Observatory; also a map of the Taieri Basin (pl. XI.)
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
|Date.||Direction of Storm Average.||Total Rainfall in Inches.|
|1867–November 19th to December 1st||N.W. to N.E.||4.002|
|1868–January 29th to February 8th||N.E. to S.W.||5.246|
|1876–January 22nd to 24th||S. to S.W.||4.960|
|1877–February 2nd to 9th||N.E. to S.W.||6.200|
|1879– March 24th to April 1st||N.E. to W.||4.728|
|" June 28th to 30th||S.E. to S.||4.560|