Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 88, 1960-61
This text is also available in PDF
(361 KB) Opens in new window
– 1 –

Ground Water Conditions in Canterbury

[Communicated by Mr. B. W. Collins and read before Canterbury Branch, July 22, 1957; received by the Editor, March 12, 1959.]


The artesian area around Christchurch has a low rainfall, and in this paper the question of how much of the available rainfall reaches the water table to replenish it has been studied from an engineering point of view. The conclusion is reached that it is only on rare occasions that rain water reaches the water table, and this seems to indicate that the Waimakariri River water keeps the sub-strata charged.

In Canterbury there is a low rainfall of 25 inches per year, but in many places there is a very high water table. It is popularly supposed that the ground water is derived from the rain falling on the surface immediately above it. This is not necessarily true, for only in the regions of very heavy rainfall does this occur. In many cases the source of the ground water is situated at some distance away. This is true especially of the Canterbury Plains and the area around Christchurch, with its artesian system. The purpose of this paper is to point out that, in areas of low rainfall and high evaporation, only on rare occasions and during winter months does any rainfall reach the water table.

In my paper “Fluctuating Levels in the Canterbury Artesian System” (N. Z. I. E. Proceedings, Vol. 37, 1951) it was shown that rainfall at the surface would raise the pressure in the substrata sufficiently to cause an immediate rise in the levels of artesian wells sunk to an underground stratum. The point is that conditions in the zone below the surface are in a very sensitive state.

From an engineering point of view, soil consists of solids, water, and air. When the water table is located below the surface there is a zone of moist soil, or it may be quite dry in dry climates. A capillary fringe exists above the water table which is approximately 12in high in sands, 3ft 6in to 4ft high in silts, and 10in high in clays. There is suction, negative pressure, or a pressure deficiency present here. Below the water table the soil is saturated. In dry climates the soil above the capillary water zone seldom reaches saturation, and any rainfall quickly evaporates or is used by plants.

The pressure deficiency is the negative pressure measured in Kg/cm2. Soil suction is measured by the pF value which is the logarithm of the pressure deficiency in grams per square centimetre. Thus by the use of logarithms the very high pressure deficiencies obtained at low moisture contents may be depicted. Hence suction and pressure deficiency are the same. The typical pF versus moisture-content graph is shown in Graph No. 1.

– 2 –
Picture icon

Adapted from Paper by G.D. Aitchison

G. D. Aitchison, in his paper “Some Preliminary Studies of Unsaturated Soils”, read before the second Australian and New Zealand Conference of Soil Mechanics and Foundation Engineering, has gathered the existing information, and he recognises six states of unsaturation: A, complete saturation; B, primary unsaturation; C, secondary unsaturation; D, partial saturation; E, modified primary unsaturation; F, modified secondary unsaturation.

In case A, which is the state below the water table, there is no air present, and there is no tendency for the water to travel from one zone to a similar zone near it. In other words, there is no pressure deficiency.

In case B, all the pores are filled with water, which is the state in the capillary zone immediately above the water table, and there is a tendency for the pore water to drain away from the largest pores. There is also a slight pressure deficiency.

In case C, the air begins to enter the pores, and there is a pressure deficiency.

Further desiccation of the soil can then give rise to states D, E. and F, until finally the water has been driven off, and soil solids and air remain.

As has been stated, an engineer regards soil as consisting of solids, water and air. This is shown diagrammatically in Fig. 1.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Moisture content is expressed as the percentage of water present to the weight of oven-dried soil. Thus w=Ww/Ws.

In the above diagram let Va = the volume of air; Vw = the volume of water; Vs = the volume of solids; V. = the total volume; Vv=the volume of voids=Va+Vw.

– 3 –

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The degree of saturation=Vw/Vv=Sr. By definition the voids ratio e=Vv/Vs, Porosity n=e/1+e, and e=n/1−n.

For saturated soils it may be proved that e=wG, where G. is the specific gravity of the soil.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

A typical value of e for a soil near Christchurch would be 0.90 at a depth of 9in. The corresponding value of n is 0.45. Therefore, an inch of rain falling on the surface of a dried soil will saturate 2½m. If there is an existing moisture content of 10%, and the soil is saturated at 20% it will saturate 4½in. The saturated moisture content would be 100×0.90/2.65 = 33.5%.

Thus, starting from dry conditions—and there are long periods of drought in Canterbury—the ground would be saturated for 9in after 4in of rain. Soil suction would immediately come into action and the moisture would invade the drier zones. The question may then be asked, how much of this water will actually drain downwards to the water table and replenish it? Also, at what stage of this downward

Picture icon

Pressure Voids Ratio Curve from Capper & Cassie.

– 4 –

passage of moisture will the pressure deficiency become zero, and the condition of equilibrium moisture content be established?

The equilibrium moisture content is established when all the voids are filled with water. When the soil is saturated, e = wG, so that the voids ratio may be calculated if the moisture content and the specific gravity are known.

At equilibrium moisture content, and the water table at the surface, the pressure at any depth z is called the effective pressure p̄. The effective pressure is equal to the total pressure p̄ (due to the weight of the water plus the soil) minus the porewater pressure, or p=p−uw.

p̄ may be proved equal to γ′z, where γ′ is the submerged density of the soil, and γ′ = γ−γw, where γ is the density of the saturated soil and γw is the density of water.

Also, γ = γs—n (γs−γw), where γs is the density of the solids, so that knowing e, n can be calculated. Thence γ and γ′ may be found.

From the above, effective pressure can be plotted against voids ratio (Graph No. 2).

Because the voids ratio decreases with depth owing to the increase of effective pressure, the saturation profile shows a greater moisture content at the surface than at a depth. This is shown in Graphs No. 3 and No. 4.

Commencing on March 13, 1957, after some days rain, in which over 4in fell, auger holes were sunk at the University site at Clyde Road, and moisture content and other samples were taken. The moisture-content profile is shown on Graph No. 3. Also plotted on the graph is the equilibrium moisture content distribution.

for the water table at the surface. Four inches of rain is well above the average rainfall at any one period in Canterbury; yet the moisture content at a depth of 2ft had only reached 10.7%. In other words, there could be no charging of the water table until the moisture content profile had straightened to the curve on the right. The profile shows a maximum pressure deficiency at 2ft, and to achieve equilibrium the water content would have to increase from 10.7% to 27.5%. According to the soil-suction and moisture-content graph, the pF value for a moisture content of 10.7% is 3.5. A great deal more rain than 4in would be needed to.

– 5 –

achieve this. For dry soil at the beginning it would require nearly 11in of rain falling in a short period to obtain equilibrium moisture content. This rainfall would saturate the soil to a depth of 24in, as far as the capillary fringe. As the average rainfall in Canterbury is 25in at the coast and 35in inland and there is a great deal of evaporation, it would seem that this state would seldom occur.

The amounts of rainfall (supplied by the Meterological Department in Christchurch) for the month of March, 1957, are shown in Table I.

Table I.
Inches (Inches)
March 1–5 Nil 0.00
6 0.81 0.81
7 0.97 1.78
8 Trace 1.78
9 0.27 2.05
10 1.85 3.90
11 0.73 4.63
12 0.02 4.65
13 0.05 4.70
—  Tests taken
14 0.13 4.83
15–16 Nil 4.83
17 0.12 4.95
18 0.01 4.96
19 0.17 5.13
20–22 Nil 5.13
23 0.09 5.22
24 0.01 5.23
25 Trace 5.23
26–28 Nil 5.23
29 0.03 5.26
30–31 Nil 5.26
April 1–10 Nil
—Tests taken

On April 10 a further set of moisture contents was taken, and the two sets of readings are shown in Table II.

Table II
13/3/57 10/4/57 18/4/57 May, 1957
(After 4.75in rain)
Depth Per cent Per cent Per cent Per cent
6in 26.4
9–11in 18.2 22.5
12in 21.4 230
15–17in 15.6
18in 17.1 16.4
24in 10.7 17.3 14.4 21.5
32in 19.6 27.6
33in 16.9 19.9 27.9
42in 23.6 26.7 26.1 27.8
Top of capillary fringe

After very little additional rain it will be seen that at the zone of the lowest moisture content—that is, 10.7% on 13/3/57 at a depth of 2ft, the moisture content on the 10/4/57 has increased to 17.3%. From the graph (No. 3) the moisture content at saturation is 27.5%. The moisture content at the surface has fallen owing to evaporation. The tendency to attain equilibrium moisture content by straightening the profile of the moisture-content diagram may be seen. Until the drier zone at 2ft has been wetted, there can be no downward flow of water to the top of the capillary fringe at 3ft 6in.

– 6 –
Picture icon

Variations in Moisture Content Profile with Water Table Present

After further rain in April, a further set of moisture contents at the site was taken. The profile followed the readings taken on 13/3/57 down to 2ft, where the moisture content had increased to 14.4%, but saturation had not been effected, although there is a tendency for the moisture content to increase at a depth of 2ft. In the winter months saturation might be achieved and the water table replenished. The above information is shown on Graph No 3.

The particle sizes of the soil were as set out in Table III.

Table III.
Sand Sizes Silt Sizes Clay Sizes
(above 0.05 mm) (0.005–0.05 mm) (up to 0.005 mm)
Depth per cent per cent per cent Classification
0–12in (humus)
14–17in 14 54 32 Silty clay
23–25in 27 61 12 Sandy silt
32–34in 23 65 12 Sandy silt
40–42in 4 65 31 Silty clay

The calculations for determining the saturated moisture content taken on 10/4/57 are set out in Table IV.

– 7 –
Table IV
Depth Natural G V Ws Vs Vv e Sat. w n Sr Liquid
M/C% c.c. gm c.c. c.c. % Limit
9–11in 18.2 2.65 26.5 36.69 13.8 12.7 0.92 34.5 0.48 52.5
15–17in 15.6 29.1 48.80 16.8 12.3 0.74 28.0 0.43 55.7 31.8
32–34in 19.6 41.4 64.90 24.4 17.0 0.70 26.4 0.41 74.2 26.8
40–42in 26.7 53.7 79.95 30.20 23.5 0.69 26.1 0.41 100 33.6

Water table at 42in.

Sr = Degree of saturation=Vw/Vv.

Aitchison states that at Melbourne, Victoria, the average annual rainfall is 30in, and the evaporation is 39in, and the soil is yellow podsolic. As a comparison, the length of the wet period is 6 months and that of the dry period 4 months. The seasonal limits for the pF value are 2.5 to 4.5, w varies from 13% to 4% at 6in depth, and from 30% to 24% at 2ft depth, with e between 0.90 and 0.80, and Sr between 38% and 12% at the surface, and 100% and 80% at 2ft. Below 2ft depth the moisture content decreases slightly, and Sr is 95%, and although it is very close to saturation, it cannot be said that there is a water table.

The two graphs, Nos. 4 and 5, which are taken from Aitchison's paper, show the moisturecontent profile in the dry season and the wet season. In the wet season there is a tendency to form a water table at a depth of 2ft, but below this point the moisture content falls off to a figure below saturation. In other words, the presence of a water table is not dependent upon rainfall.

From the above it is evident that rainfall is a very uncertain source of supply to the water table in lowrainfall areas such as Canterbury, and that the causes of the replenishment of the water table must be looked for elsewhere. In the Canterbury case there is the Waimakariri River running over gravel strata, which would keep the underground system charged. The level at Halkett, which is 15 miles from the mouth, is 300ft above sea level, so that artesian conditions could prevail. Professor Speight, a former Professor of Geology at University of Canterbury, always maintained this to be true. In his paper “A Preliminary Account of the Geological Features of the Christchurch Artesian System” in the Transactions of the New Zealand Institute, Vol. 43, 1910, he stated that in his opinion the Waimakariri River was responsible. Even in the drier years the Avon River continues to run without any significant change in level.

The North Canterbury Catchment Board has taken gaugings of the Waimakariri River after stable conditions of weather at the gorge outlet from the mountains, and at the bridge, which is a few miles from the mouth of the river. The readings were 1,700 cusecs at the gorge, and 1,200 cusecs at the bridge. Allowing for 100 cusecs used in water races, 400 cusecs must have entered the underground strata. When it is remembered that the volcanic mass of Banks Peninsula acts like a plug, and that deposits of the finer silts and clays at sea would cause a further barrier to flow, it may soon be realised that the conditions are ripe for the storing of water

– 8 –

underground. Of the 400 cusecs which finds its way underground under these conditions, only about 50 cusecs are used for water supply in the area.


An amount of 4.75in of rain fell between May 16 and May 20, 1957, making the total for nearly five months 17in. As this is well above average, one would expect the water table to approach the surface. This occurred in many places, but at Clyde Road moisture-contents samples still showed a deficiency. At a depth of 2ft the moisture content was 21.5 per cent, saturated conditions being about 27 per cent.

If the available space in the voids of the soil is calculated as it existed on 18/4/57, it is found that it will take exactly 4in of rain to bring the profile of moisture content out to the saturation stage.

Some subsoil drainage tests have been carried out with a lysimeter which show that there is a downward tendency for the flow. Any apparatus which destroys the natural conditions which exist in the subsoil should not be used.


Terzaghi, K., and Peck, R. T., 1948. Soil Mechanics in Engineering Practice New York: John Wiley & Sons, Inc.

Capper, P. L., and Cassie, W. F, 1948. The Mechanics of Engineering Soils, Second Edition. Billing & Sons Ltd., Great Britain.

Aitchison, G. D., 1956. Some Preliminary Studies of Unsaturated Soils. Proceedings of the 2nd Australian and New Zealand Conference on Soil Mechanics and Foundation Engineering.

Baver, L. D., 1948. Soil Physics. New York: John Wiley & Sons Inc.

Northey, R. D., 1956. Soil Moisture and Civil Engineering. Proceedings of the Conference on Soil Moisture at D. P. L. Wellington, N.Z. Department Scientific and Industrial Development Information Series.

Oborn, L. E., 1955. The Hydro-Geology of the Canterbury Plains Between the Rakaia and Ashley Rivers (Thesis).

— 1956. Ground Water in Metropolitan Christchurch. .D.S.I.R. Hydrological Report, 128.

P. J. Alley

, B.E., A.M.I.C.E., M.N.Z.I.E., M.I.S.S.M.F.E.,
School of Engineering,
University of Canterbury,