Art. LXVIII.—On the Behaviour of Two Artesian Wells at the Canterbury Museum.
[Read before the Philosophical Institute of Canterbury, 3rd July, 1895.]
Previous to March, 1894, the Museum was supplied with water from a well sunk into the first water-bearing stratum, and probably between 90ft. and 100ft. in depth. This failing to give a sufficient supply, a second well was sunk, which reached the second water-bearing stratum, at a depth of 190ft. from the surface, on the 23rd March, 1894. A glass gauge, connected with this well, was put up inside the Museum on the 9th April, and the height at which the water stood above
the stone sill of the back door of the Museum was registered: this was 9ft. 8 ½in. On the 11th it went down to 9ft. 6in., and on the 21st it had risen to 9ft. 8 ¾in. This I call the “deep well.”
The top of the iron pipe of the old well—or shallow well—I found to be 2ft. 10in. above the stone door-sill, and the water stood 2in. or 3in. below the top; so that there was a difference of about 7ft. in the height of the water in the two wells.
At this time the rainfall was not registered in Christchurch, but arrangements were being made to put up a gauge in the Public Gardens; and Mr. A. L. Taylor, the head gardener, has kindly supplied me with the rainfall from the 1st June, 1894, to the 31st May, 1895.
Meantime, as it was known that the artesian wells near New Brighton were affected by the height of the tide, I made hourly observations on three different days to see whether the Museum wells were also affected; for, if this were the case, a tidal correction would have to be applied to every observation. I found, however, that the tide had no appreciable effect on either of the wells.
At first I had some little difficulty in reading the deep well on account of the hydraulic ram, which would occasionally jamb; but I remedied this by stopping it altogether and keeping it stopped during the whole period of the observations. These observations were made between 10 a.m. and 11 a.m. each day. As the glass gauge reached only to 10ft. above the door-sill, all observations above that level could only be conjectural, and I gave up making them. When the water rose above the top of the iron pipe in the shallow well I made a cardboard collar to lengthen it. On account of leakage these readings are only approximate, but they are fairly accurate.
The results of the observations I give in a series of twelve diagrams, one for each month, on which the daily rainfall in the Public Gardens is also shown. In these diagrams the figures in the right and left columns are the number of feet and inches from the datum-line on the door-stone, and to read the height of the deep well 7ft. must be added. Thus, on the 1st June, 1894, the height of the shallow well was 3ft. 1 ½in., and that of the deep well 10ft., above the door-sill. The daily fluctuations of the shallow well are represented by a continuous line; those of the deep well by a broken line. The rainfall is shown in the usual way at the bottom of each diagram. The figures at the top and bottom of the diagram are the days of the month, and the thick longitudinal lines divide the month into weeks.
I will first take the connection between the rainfall and the height of the water in the wells:—
June, 1894.—May, 1894, closed with heavy rain, and on 1st June both wells stood very high; but, although slight rain continued at intervals until the 10th, both wells had fallen considerably by the 15th. Heavy rain on the 24th made them rise again.
July, 1894.—During the first week in July the wells again went down; but the rain on the 8th and 10th, although not heavy, made them overflow and sent them up again to a height perhaps not much less than they had during the first week in May. By the 24th they were down again. It is remarkable that the rain which fell on the 8th and 10th May, although only 0.17in. more than that which fell on the 24th June, should have sent the wells up so much more. Two explanations suggest themselves: one is that the rain of the 24th June fell more rapidly and, in consequence, ran mostly off the surface; the other is that the rain of the 24th June was local, while that of the 8th and 10th July was more widely spread. I have not sufficient data to determine which of these two explanations is the more probable. The rain of the 25th July again raised the wells slightly, but by the end of the month they had both fallen to lower levels than any they had stood at during the two months.
August, 1894.—The rains of the 2nd and 3rd August again raised the wells,* but by the 8th they had again fallen. Continuous light rain from the 14th to the 19th raised them once more, but not so high as might have been expected. After this they sank, and before the end of the month had attained the average level.
September, 1894.—Continuous light rain fell during the whole of the first week. Both wells again overflowed, and did not come down to the average level until the 18th. Heavy rain on the 24th and 25th again sent them up, and they continued high until the end of the month.
October, 1894.—Rain on the 1st and 2nd kept the wells up; but they gradually fell during the rest of the month—there being no rain—until at the close they stood lower than they had done before, and the average level of June to October was never again maintained during the rest of the year.
November, 1894.—In this month the rain was fairly well distributed, and the wells kept at an average which was 3in. below that of June to September, the drop from one to the other taking place during the dry weather in the middle of October.
[Footnote] * The defect in the record of the deep well at this time is owing to the ram having gone wrong.
December, 1894.—In this month the rainfall was very small, and the wells fell another inch. Towards the end of the month another fall of about ½in. occurred.
January, 1895.—This month opened with rain, which, however, did not affect the shallow well, although the deep well rose. The next week, from the 6th to the 11th, was rainy, and both wells went up, falling again on the 16th. Showery weather once more raised them a little; but after the 20th they both fell, and—especially the deep well—went down to levels not before reached.
February, 1895.—Slight showers between the 3rd and 7th did not affect the wells, which continued to go down until the 20th or 22nd, when the rain raised them a little; but they had fallen again by the 27th.
March, 1895.—There was practically no rain during the first fortnight of this month, and the wells fell slightly until the 20th, when four days of rain sent them up again to the level they had at the end of January.
April, 1895.—During the first week of this month the wells fell slightly; but rain occurred on the 8th and again from the 12th to the 16th; the wells rose to the level of the middle of January, and continued at this level notwithstanding that there was no more rain.
May, 1895.—There was but little rain during this month; nevertheless the wells did not fall, but kept steadily the average of the latter part of April.
The following table gives the average height of the wells and the rainfall for each month:—
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
|—||Shallow Well.||Deep Well.||Rain.|
|1894.||Ft. in.||Ft. in.||In.|
|May||2 7 ½||9 8||(?)|
|June||2 10||9 10||1.84|
|July||2 10||9 10||2.09|
|August||2 10||9 10 ½||2.51|
|September||2 10 ½||9 10 ½||3 46|
|October||2 9||9 9||0.50|
|November||2 7 ½||9 7||2.57|
|December||2 6||9 5||3.30|
|January||2 5 ½||9 5 ½||2.23|
|February||2 4 ½||9 2||1.21|
|March||2 4 ½||9 2||2.44|
|April||2 5 ½||9 4 ½||1.71|
|May||2 5 ¼||9 5 ¼||1.35|
It thus appears that both wells are very sensitive to rain, rising rapidly and falling again; but not so rapidly as they
rise. Exceptions are the rain of the 1st and 2nd January, which did not affect the shallow well although the deep well rose, and the rains of the third week in November, which had no effect on either of the wells. In both these cases several weeks of drought had preceded the rain, and it seems probable that the dry ground soaked up the water and allowed none to penetrate to the reservoirs connected with the wells.
But, although the wells rise with rain, they also rise without any rain being recorded. This may, perhaps, be sometimes due to rain falling outside Christchurch—especially in cases like those of 25th July, 27th November, 12th December, 16th and 27th March, and 9th May, when the shallow well rose rather before rain fell in Christchurch. On the 26th December heavy rain fell at Springfield and the Malvern Hills, although none fell at Christchurch. Both wells fell that day, but next day the deep well rose slightly, although the shallow well still fell.
Far more important is the periodical rise shown on Sundays and holidays, which cannot possibly be due to rain, and which is specially noticeable in the rise of the shallow well every Sunday. This well really rises every night, and on Sunday morning does not fall as it does on week-days. The rise varies from ¼in. to 1 ½in., and the average is rather more than ¾in. The same occurs with the deep well, but it is masked by the great and rapid fluctuations which take place in this well. These, as recorded on Sundays, vary between - 1in. and +4in; but here also the average comes out about ¾in. rise. On the Prince of Wales' Birthday (9th November), Christmas Day (Tuesday), and Good Friday (12th April), both the wells rose. On the Queen's Birthday (24th May) the shallow well rose and the deep well fell. On the anniversary of the province (Monday, 17th December) both wells fell, just as they generally do on a Monday. Easter Monday was rainy, and the wells rose with the rain. This periodic rise must depend on a constant supply of water, either from some lower-lying water-stratum or from the bed of the River Waimakariri.
I have been unable to get complete information about the floods which took place in the River Waimakariri during the year, but I kept a record of the north-west winds, and these are generally followed by floods; and, as these are the only floods unaccompanied by rain, they are the only ones which can be used to ascertain if the level of the wells is affected by the height of the river.
Hot north-west winds occurred on the following days: 19th October, 9th and 10th November, 13th November, 24th November, 1st December, 17th December, 7th, 8th, and 9th
March, 15th March, and 22nd April. Of these the hot winds of the 1st and 2nd December, and of the 7th, 8th, and 9th March, caused heavy floods in the river sufficient to be noticed in the newspapers. The hot winds of the 13th and 24th November and of the 22nd April were accompanied by rain, and must therefore be omitted. With the other seven it did not rain, but that on the 9th November is complicated by the holiday. With reference to the others, the diagrams show that the wells fell on the 19th and 20th October and on the 10th and 13th November. On the 24th they were stationary. On the 1st December they fell. During the heavy floods of the 2nd and 3rd the rise on the 2nd was no doubt due to that day being Sunday, for on the 3rd the deep well remained steady, and the shallow well rose slightly. On the 5th and 17th December both wells fell; on the 7th March they rose; on the 8th the deep well remained steady while the shallow well fell; and on the 9th the deep well rose and the shallow well fell. On these two last days the river was in high flood. On the 15th March the deep well rose and the shallow well remained steady. It follows therefore that the wells do not rise with the river, but are apparently unaffected by it.
Up to the middle of January, 1895, the two wells fluctuated fairly well together, the average difference in height being 7ft. But on the 22nd January the deep well fell far more than the shallow well, and this relatively lower level was maintained until April, when the two wells nearly resumed their old relations, the deep well being about 6ft. 11in. higher than the shallow well. Possibly this may have been due to the water from the deep stratum being used for watering the streets during the hot months.
Some of the vagaries of the deep well between the 12th and 29th November may be due to large quantities being used for washing moa-bones. During the latter part of October and again in December the height of the water in the deep well was very variable. Great falls occurred between the 14th and 20th December, and between the 23rd January and the 2nd February, and a third between the 16th and 21st February. For these I cannot account.
The conclusions which seem to be justified from the observations of so short a period as one year are as follows:—
1. The deep well fluctuates more than the shallow well. This is, no doubt, partly due to the far greater number of wells sunk into the first stratum, all of which would rise and fall together; consequently it would require more water to raise them all an inch than if there were fewer of them. But it may be partly due to the head of water above the level of Christchurch being smaller in the deep stratum than in the shallow stratum.
2. Many fluctuations, apparently erratic, are no doubt due to irregular causes of which I can give no account—such as the stopping of rams; less pumping by windmills on calm days; rainfall outside Christchurch; watering the streets; opening of new wells, &c.
3. Certain fluctuations are sufficiently regular to show cause and effect: they are—(a) Rise after rain, unless the ground is very dry; (b) rise at night, owing to less water being used.
4. The rise at night in each well is about equal; consequently, as the second stratum fluctuates more rapidly than the first stratum, the supply to the first stratum must be larger than that to the second.
5. This rise without rain appears to be the main source of supply. It must be due either to water running in from the bed of the River Waimakariri, or to some still lower water-bearing stratum leaking into the upper ones.
6. The height of the water in the wells is not affected by the height of the river. It is therefore improbable that the supply is from a leak in the river-bed.
7. The regular inflow—independent of rain—is, with each stratum, greater than the present outflow by night, but smaller than the outflow by day.
Lastly, we come to the question of the annual rate of lowering of the water. Of course, one year's observations are not enough to settle this point accurately, but they are better than nothing. The shallow well stands about 1 ¾in. and the deep well about 2 ¾in. below the level of last year. The annual rainfall was 22.21in., which is 2in. or 3in. below the average; but I do not think that this will account altogether for the fall. No doubt heavy rain would make the wells go up again to last year's level, but they would go down again in two or three weeks.
The shallow wells are estimated by Mr. T. Danks to have fallen between 6ft. and 7ft. during the last thirty years, which gives an average annual fall of between 2 ½in. and 2 ¾in. Mr. H. Oakley has estimated that they have fallen 3ft. in twenty years, which is rather more than 1 ¾in. a year. Mr. Oakley also thinks that the deep wells have fallen 3ft. in six or seven years, which is an average of 5 ½in. a year. This rate of lowering requires further investigation; but no doubt it is serious, and increasing year by year.