Other formats

    TEI XML file   ePub eBook file  


    mail icontwitter iconBlogspot iconrss icon

Geology of the Provinces of Canterbury and Westland, New Zealand : a report comprising the results of official explorations

(C) The Formation of the Canterbury Plains

(C) The Formation of the Canterbury Plains

This chapter would be incomplete were I not to offer a few notes on the formation of the Canterbury plains, about which so much has been written, and so many theories have been brought forward, that a matter of apparently so much simplicity has in course of time become quite obscured. Since my report on the formation of the Canterbury plains was published in 1864, all the levels, surveys, engineering works, together with well-sinking, have amply confirmed my views that the Canterbury plains are of fluviatile origin, that, with the exception of some morainic accumulations in the upper portion and the drift sands round Banks' Peninsula, and the partial lacustrine deposits filling the former extension of Lake Ellesmere, the whole of the plains were formed by the deposits of huge rivers issuing from the frontal end of gigantic glaciers. Mr. J. T. Thomson, the Surveyor-General, in a paper on "The Glacial Action and Terrace Formation of South New Zealand," published in Vol. VI of "The Transactions of the New Zealand Institute," gives a diagram in illustration of the fan theory on page 329, and alludes to the sluice page break page 397deposits, such as they are formed at the present day on flats like Gabriel's Gully and Weatherstone, conforming in every respect on a small scale with the deposits formed at the termination of the gigantic glaciers during the Great Ice period. Professor Hutton in his various writings has been contending that the Canterbury plains must be of submarine origin, and that afterwards they have been raised above the sea level, when the rivers excavated their present channels, as terraces could only be formed during a rising of the land, and he brings forward a formidable array of scientific authorities in support. However, similar terraces, as I have pointed out previously, occur in all Alpine regions, in the European Alps, the Himalayas, the Rocky Mountains, and many other localities, where in several instances there is overwhelming evidence towards proving a subsidence of the lands since post-pliocene times. I should not have alluded to this question did not Professor Hutton in his last work, "Geology of Otago,"* still persist in claiming for the Canterbury plains a submarine origin, although he adopts for the former province the theory that the land during the Great Glacier period stood at a higher level than it possesses at present. As according to him, river terraces prove elevation, he fails to explain how the beautiful and perfect series of terraces in the Clutha valley could have been formed during subsidence, and it is thus difficult to reconcile his statement, that the plains in Otago "have certainly been formed in the way suggested by Dr. Haast." I wish here once more to repeat that the Canterbury plains on either side of the Malvern Hills are not on the same level; on the contrary, the difference is very considerable, and as the railway and other levels have shown, the fans of each river are quite distinct. The sections of the Canterbury plains attached to this Report, Nos. 1 and 2, on plate 8, drawn from more correct information than I formerly possessed, will, I trust, settle this question definitely. The diagrams in Mr. Thomson's excellent paper, previously mentioned, giving the curves of several valleys and the theoretical bearings of the question at issue, further confirm my views, that the Canterbury plains have been exclusively formed by the accumulations of the rivers still flowing through them.

Advancing now to a consideration of the general laws by which the formation of river beds are regulated, first, as far as I am aware,

* "Geology of Otago;" Hutton and Ulrich. (page 91).

page 398proposed by Italian engineers and natural philosophers, the following formulæ will conclusively show that all the principal physical features can be explained by them:—
  • 1 "The more a river advances from its sources, the less will be the declivity of its bed." Or stated differently—"All rivers dispose their beds in the lower course on a less slope than they had in the upper, so that the declivity diminishes in proportion to the distance they have to run from their sources." This law holds good for slowly flowing rivers, depositing ooze only, as well as for rivers and torrents, rolling boulders, gravel, and shingle, because the largest material is gradually left behind.
  • 2 "The greater the ordinary body of water is in a river, the Jess will be the slope of its bed." Or, stated in other words—"The slope of the bed of a river will diminish in the same proportion in which the body of water is increased."

These two natural laws will be quite sufficient to explain all the principal phenomena which our rivers exhibit, the more so as none of them south of Banks' Peninsula have formed a delta properly speaking. The only river possessing such complete features is the Waimakariri, which in its lower course shows great resemblance with the lower part of the Po, in the Plains of Lombardy, which for many centuries has caused anxiety for the safety of the surrounding country, its bed having been raised gradually by embankments to such an altitude that it runs in many localities above the level of the plains on both sides.

We must conclude that the Canterbury plains are formed by the outlets of enormous glaciers, large torrents bringing down with them the morainic matter, thrown in their course at the terminal face, raising their beds and shifting their channels at the same time so as to form fan-shaped fluviatile accumulations, consisting of shingle, gravel sand, and glacier mud. In applying the preceding rules, we shall find that as the sources of the post-pliocene torrents were at the terminal face of the glaciers, they lay much further to the east than the present ones, and at the same time, those glaciers being on such a gigantic scale, the torrents issuing from them must naturally have been so much the larger. In consequence we ought not to feel any astonishment at observing, if we take into account both circumstances of the retreat of the glaciers combined with their diminution, that the rivers cut new courses into the older deposits in recent times, the page 399more so when we consider their volume was so greatly reduced. It is thus now easily understood that, for instance, the post-pliocene fan of the river Ashburton was so enormous, and that as the then glaciers, all assisting in the formation of these large deposits, shrunk back to their proper valleys, the outlet from the one remaining Ashburton glacier does not stand in any relation with its former fan. Consequently it has not been able to lower its channel to any extent, whilst the Rakaia, which is still the outlet of considerable glaciers, and has retreated much more than the former, has been able to cut its channel much deeper, and to prepare a much more uniform gradient for its bed.

As the mountains began to become gradually eaten into by the action of the descending ice-masses, sharper ridges and peaks were formed, so as to lessen the extent of the surface where perpetual snow could accumulate to feed those glaciers, which consequently began to retreat. Therefore the sources of the rivers which now cross the plains were lying, in the post-pliocene epoch, much nearer to the sea, their fans were much steeper, and they continued to build them higher and higher, changing their apex continually, and at the same time forming with their lowest portion the shore line of the ocean.

All my observations show at the same time that the three great post-pliocence torrents, namely, the Ashburton, Rakaia, and Rangitata were united without doubt in one large water-course before they reached the sea, having a common fan at their mouth, following the law that rivers which unite, endeavour to do so by the shortest line. This fact is well illustrated near the present mouth of the river Hinds. This river flowing between the fans of the Rakaia and Ashburton above the point where they united into one, has been forming a swamp of considerable extent, being four to five miles broad and eight miles long. It has partly been drained by natural water-courses, partly by artificial drains under process of formation, all passing through the higher shingle beds in front. The united existence of such a fan is also instructively shown at the coast, where the sea has worn away the protruding arc between the southern bank of the Rakaia and the northern bank of the Rangitata. These cliffs, consisting of true river shingle and river sand, are at the south side of the Rakaia about six feet high, and rise gradually to an altitude of 70 feet near the mouth of the Wakanui Creek, four miles north of the Ashburton river, where consequently the united fan has suffered the greatest destruction. Having reached this altitude the coast cliffs page 400towards the south diminish in height, so that near the mouth of the Ashburton they only rise to 18 feet.* A few miles north of the Ashburton we reach again the line, where the sea-coast meets the level of the fan under consideration.

To understand fully the combined action of the agencies at wort, it will be necessary to remember, that the largest fan was necessarily formed by the most extensive glacier, and in consequence the fan of the post-pliocene Waimakariri could not reach so far into the sea, as did the united Ashburton one, built up by the southern rivers. In adopting the present data before us as the basis of reasoning, we shall find that, as before observed, the united southern fan began near the mouth of the Rangitata, advanced into the sea till it attained a breadth greater by several miles than that which it now possesses near the mouth of the present Wakanui Creek, and crossed again near the mouth of the Rakaia, of which a portion has been, and is now being, washed away. Continuing the same arc of this fan in a north-west direction we shall reach the Selwyn several miles above its junction with Lake Ellesmere, where the fan of the Waimakariri joined it, and, in consequence, the Selwyn, like the Hinds, is flowing upon the junction of both fans. During this post-pliocene era a narrow arm of the sea ran along the western foot of Banks' Peninsula, as shown by the drift sands and raised beaches surrounding it here. The oceanic swell south of Banks' Peninsula travelling towards the north, assisted by the action of the waves, very soon began to disintegrate the post-pliocene accumulations. The shingle and sand derived from that destruction, travelling northwards, were augmented considerably by the material of the same nature brought down by the rivers, and, assisted by the two prevailing winds, finally formed a dam from the mouth of the Rakaia to that isolated volcanic system, becoming every year more considerable. A peninsula was soon originated, the connecting isthmus being that dam or broad shingle spit behind which an arm of the sea formed a bay, moderately deep and comparatively sheltered, with an entrance to the north of Banks Peninsula. Into this bay the Selwyn, Waimakariri, and probably, at one time, the Rakaia fell. Gradually the shingle and sand brought down by the northern rivers, the Waipara and Ashley, travelling with a northerly swell towards the south, threw, in their turn, a bar across the opening of the bay, thus forming a large

* From observations of Mr. C. W. Adams, Geodetical Surveyor, kindly furnished to me.

page 401lagoon, of which Lake Ellesmere is only a remnant. Of this occurrence we have sufficient evidence in the immediate neighbourhood of Christchurch and Kaiapoi; and still up to the present time the topographical features of the ground which rises close to the western foot of Banks' Peninsula, only about twenty-six feet in its highest neck above the sea, having on both sides low swampy ground, would give additional evidence to such a theory, were not other proofs sufficient.

I have already pointed out that there has been an oscillation of about twenty feet, the upward movement occurring towards the end of the Great Glacier period. This rising of the country was of course of great importance for the low ground forming in front of the shingle fans. Thus with the rising of the country this large lagoon became partly filled either by the shingle of the rivers falling into it, (of which the Waimakariri was the most important in throwing its fan nearly across it), or assisted by the silt which the rivers had in suspension, and which was thrown down all over the lagoon in the form of loam and clay. Moreover, drift sands invaded it from its eastern shore, coming from the north, giving additional breadth and stability to the banks which separated it from the Pacific Ocean. At favourable localities, just rising to or near the surface of the water, vegetation began to spring up, forming swamps, which with the assistance of bog-mosses (Sphagnum) and other aquatic or semi-aquatic plants forming peat, raised considerably the ground, so as to form a suitable locality for the beginning of forest vegetation. Many were the slight oscillations taking place, through which the rivers changed their channels, and either removed the clays deposited in the lagoon, or buried below their newly-forming shingle-beds the peat swamps, vegetable soil, and even forests, of which well-sinking in the neighbourhood of Christchurch has offered many illustrations. In this way those swamps were either covered or partly drained, whilst others, not so favourably situated, continue to this day to remain in the same condition, as, for instance, portions of the Rangiora swamp. Many of the raised beaches are easily traceable, and with them the former western banks of the great Lake Ellesmere extension. This accounts also for many peculiarities we meet in the centre of this ancient, partly drained and filled lagoon, which, without such an explanation, would be unaccountable.

It has been before stated, that the declivity of a river diminishes in proportion to the distance it has to run from its source, and it is thus natural that the rivers now traversing the Canterbury plains have page 402lessened their fall, although still having the character of true torrents, descending in their course at a more uniform but lesser rate than the post-pliocene fans of the former larger rivers. They have, in consequence, according to their present size and position of sources, cut more or less deeply into these deposits, but reach a point where they intersect the line, below which the old fan falls more considerably, and thus the present rivers, instead of excavating, begin to fill up, raising their beds above the older deposits and forming new fans. The present rivers repeat simply on a smaller scale the action of the large post-pliocene torrents forming the upper plains; they also debouch from a gorge, although the sides of it consist only of shingle, and spread fan-like over the lower grounds, the axis of the course of the river changing so as to build it up regularly. Consequently it has been found by actual survey, that the contours taken at an equal distance from the beginning, or perhaps better stated, from the emergence from the older beds, have an equal gradient and altitude. Thus, as before said, the higher this point lies above the junction of the river with the sea, the larger will be the radius over which the river can roam and raise its bed. It was, therefore, of the greatest value to the district to ascertain not only the exact spot where the recent accumulations begin to spread over the older ones, but also to find, by careful measurements, what is the ratio of the present rise above the level of the sea, as well as above a given spot of the plains of older origin. The fact, however, that only the Waimakariri, and in a minor degree the Rangitata on its southern and the Rakaia on its northern banks, for a small extent are subject to such an occurrence, is worth recording, and shows distinctly that the two last-mentioned rivers flow on the sides of the large post-pliocene deposits described previously. Here some rich alluvial land is situated, and thus those small strips, endangered by the aberrations of these rivers, are of far greater importance than a look at the map at first might suggest.

In the following four tables the data collected during a number of years, as to the fall of the plains, and the rivers which formerly built them up, are given. They are taken from my Report on the formation of the Canterbury plains, to which a number of details obtained since and also some corrections have been added.

page 403
Table I
Showing the Fall of the Canterbury Plains.
Name Of Rivers. —— Difference between two Stations. Length of Plains. Fall of Plains Per Mile.
Feet. Miles. Feet.
Rangitata From beginning of gorge to mouth mean 1443 32 45
From beginning of gorge, 1443 feet, to railway crossing 292 feet 1151 23¼ 49½
*From railway crossing 292 feet to sea 292 33⅓
Ashburton From Two Brothers to sea mean 1500 35½ 42¼
From Two Brothers 1500 feet, to railway crossing 305 feet 1195 25 48
*From railway crossing 305 feet to sea (50 feet above sea level) 255 10½ 29
*Plains between Rakaia and Ashburton (railway crossing), altitude above sea 412 feet, cliffs 60 feet 352 12½ 28
Rakaia *From gorge to sea mean 1480 37½ 39½
*From gorge 1480 feet, to railway crossing 378 feet 1102 21½ 51¼
*From railway crossing 378 feet to sea 378 16 23½
Waimakariri From upper gorge to sea mean 1580 44 36
From junction of Kowhai 1410 feet, to lagoon at gorge hill 1182 feet 228 5 41½
From lagoon 1182 feet, to the so-called 18th mile peg 355 feet 827 18¾ 44
*From 18th mile peg 355 feet, to last raised beach near North-road 33 feet 322 13¼ 24¼
*From last raised beach 33 feet to sea 33 4 8
Selwyn From entrance into plains to Lake Ellesmere 790 29 27¼
From entrance into plains 790 feet, to railway crossing 214 feet 576 16½ 35
*From railway crossing to Lake Ellesmere 214 12½ 18

* The numbers to which an asterisk has been prefixed are results obtained with the spirit-level, by the Public Works Department.

page 404
Table II.

The length of the Rivers is taken by following their general direction.

* The numbers to which an asterisk has been prefixed are results obtained with the spirit-level, by the Public Works Department.

Showing the 'Fall of the Rivers of the Canterbury Plains, from their sources to the Sea, and their intermediate gradients.
Name of Rivers. —— Difference between two Stations. Length of Rivers. Fall of Rivers Per Mile.
Feet. Miles. Feet
Rangitata Whole course from the Clyde Glacier main source 3762 feet to sea mean 3762 75 50
From Clyde Glacier 3762 feet, to junction with McCoy 3269 feet 493 3 164⅓
From junction of McCoy 3269 feet, to junction of River Havelock 2192 feet 1077 11 98
From junction of Havelock with Clyde 2192 feet, to beginning of Canterbury Plains 1180 feet 1012 29 35
From beginning of Canterbury plains to railway crossing 257 feet 923 23¼ 39¾
*From railway crossing 257 feet to sea 257 29
Ashburton Whole course from Ashburton Glacier 4823 feet to sea mean 4823 67 72
From Asliburton Glacier 4823 feet to Camp Creek 3717 feet 1106 4 276½
From Camp Creek 3717 feet, to junction with Clearwater Creek 1832 feet 1885 16 117¾
From junction of Clearwater Creek to beginning of plains, near Two Brothers 1400 feet 432 11½ 37½
From Two Brothers 1400 feet, to railway crossing 300 feet 1100 25 44
*From railway crossing 300 feet to sea 300 10½ 28¾|
Rakaia Whole course from Ramsay Glacier 3354 feet to sea mean 3354 85 39½
From Eamsay Glacier to junction of Cameron 2034 feet 1320 13 101½
From junction of Cameron 2034 feet to junction of Mathias 1688 feet 346 6 57⅔
From junction of Mathias 1688 feet, to junction of Wilberforce 1357 feet 331 35
From junction of Wilberforce 1357 feet, to Gorge Island 875 feet 482 19 25½
*From Gorge Island 875 feet, to railway crossing 372 feet 503 21½ 23½
*From railway crossing 372 feet to sea 372 16 23¼
Waimnakariri Whole course from Waimakariri Glacier 4162 feet to sea mean 4162 93 44¾
From Waimakariri Glacier to junction of two main source branches, [unclear: 2607] feet 1555 5 311page 405
Waimakariri From junction of two main source branches to junction of Crow River 2273 feet, 334 4 83½
*From junction of Crow River to junction of Bealey River 2065 feet 208 5 41¾
From junction of Bealey to junction of Esk River 1562 feet 503 21 24
From junction of Esk River to junction of Kowhai River 1003 feet* 559 17 33
*From junction of Kowhai River to White's old Accommodation House in river bed 605 feet 398 15 26¼
From White's old Accommodation House 605 feet, to tidal boundary 605 22 27½
From near the crossing of the North-road to sea 4
Selwyn From entrance into plains to Lake Ellesmere mean 767 29 26½
From entrance into plains 767 feet, to railway crossing 212 feet 555 16½ 33½
*From railway crossing 212 feet, to Lake Ellesmere 212 12½ 18
Table III.—Showing the Fall of the River Waitaki.

For comparison I may offer the principal data in my possession of the main branch of that River, and after its junction with the other branches of the Waitaki to its mouth.

Tasman From the Tasman Glacier 2456 feet, to Lake Pukaki, 1717 feet 739 25 30
From outlet of Lake Pukaki to mouth of Waitaki 1717 82 21
Waitaki If Lake Pukaki should be filled up by the River Tasman, the River Waitaki would show the following results 2456 117 21
The beginning of the old post-pliocene fan of the River Pukaki, the main branch of the Waitaki, is situate 150 feet above that river at the lake, which shows that the former outlet of that enormous glacier would have had only a fall to the sea of 1967 82 23

As for that river, the last three miles before it enters the Lake is almost without any fall, it flowing through swampy deltaic ground, I have calculated two miles less for its course than it actually has, so as to give a more correct result of its gradient.

page 406
Table IV.

In order to prove that the larger the river, the greater its power of scouring its channel, and consequently of lowering its bed, the following data will not be superfluous. The rivers follow according to their volume.

Rivers falling per mile, in feet:—
Waitaki. Rakaia. Waimakariri. Rangitata. Ashburton.
From, sources to sea 21 39½ 44¾ 50 72
From beginning of plains to sea 23 23½ 28 37 39¾
Table V.

The following data will show the difference in the fall of the post-pliocene rivers, from, the beginning of the plains at the foot of the mountains to the sea, for the calculation of which the surface of the Canterbury Plains offers us the necessary data:—

Rakaia, 39½. Waimakariri, 36. Rangitata, 45. Ashburton, 42½.