Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 53, 1921
This text is also available in PDF
(3 MB) Opens in new window
– 131 –

Art. XVI.—The Warped Land-surface on the South-eastern Side of the Port Nicholson Depression, Wellington, N.Z.

[Read before the New Zealand Science Congress, Palmerston North, 28th January, 1921; received by Editor, 1st February, 1921; issued separately, 4th July, 1921.]


The Problem 131
   The South-eastern Boundary of the Depression 131
   General Tectonic Features of the Depression 133
The Evidence 134
   Summary 134
   Tilted Coastal Platforms 135
      The Platforms farther West 135
      Platforms of the Eastern Side tilted towards the Depression 136
   Evidence from Drowned Valleys 138
   Evidence from Regraded River-valleys 140
Appendix: The Problem of the Turakirae Coastal Plain 142
List of Papers referred to 143

The Problem.

In 1912 the writer described Port Nicholson, the harbour of Wellington, as occupying an area of subsidence with somewhat indefinite boundaries, and later, in 1918, in a brief account of the coastal features of New Zealand, termed it “a locally downwarped and embayed area.” It had previously been described by Bell (1910) as a complex graben. The present article is concerned with evidence of warping on the south-eastern side of this area, which is here termed the “Port Nicholson depression”—warping that is of interest not only from a geological point of view, because of its sharpness, but also from the viewpoint of geography, since it forms a boundary of one of the finest natural harbours in the world (fig. 1).

The South-eastern Boundary of the Depression.

Bell appears to have regarded the eastern boundary of the Port Nicholson depression as a fault-scarp, and eastward of the harbour—between it and the Rimutaka Range—his map and profile indicate the presence of a narrow fault-bounded block standing lower than the block forming that range (1910, pp. 537, 539). No mention of tilting of this step-like intermediate block was, however, made by him.

The writer, in 1912, inclined to the belief that either the original boundaries of the subsided block were flexures rather than faults, or, on the other hand, the original subsidence had taken place so long ago that topographic evidence of faulting had been destroyed. The distinct and prominent scarp of the Wellington fault along the north-west side of the harbour was regarded as of much more recent origin than the depression as a whole, which, though it must have been deepened, was not initiated by the late insinking along this fault. Field-work having been mainly confined to the area on the western side of Port Nicholson, the question of the eastern, or south-eastern, boundary of the whole depression was then left open (p. 262), but reasons were given for rejecting the fault-scarp

– 132 –

theory in explanation of it (pp. 261–62). It was pointed out that the slope descending to Port Nicholson is maturely dissected, and that the only facet-like forms and blunt-ended spurs more or less in line occur where cliffing by marine erosion has been recently in progress, and this slope as a whole was ascribed without hesitation to the work of normal erosion guided by structure, it being an erosion-scarp along the outcrop of a resistant highly-inclined stratum.

It may be added that the eastern side of the harbour is now distinctly a shore-line of submergence (Plate XXIX, fig. 1) modified by marine erosion, especially at the southern end, where waves driven into the harbour-entrance by southerly winds still retain considerable energy and have cut a continuous line of cliffs with a height in places of 300 ft. The embay-ments produced by the submergence are small, but this is because of the steep declivities of the drowned ravines which dissect the erosionscarp. Similar embayments, filled with alluvium, are present farther north-east along the edge of the Hutt River delta, which partly fills the Port Nicholson depression (Plate XXXIII, fig. 1), and also still farther north-east, where the depression widens again after a constriction is passed, and is occupied by a basin-plain in which lie Trentham and Upper Hutt (Plate XXXIII, fig. 2).

Picture icon

Fig. 1.—Locality map of the Port Nicholson area, New Zealand.

The coincidence of the trend of the erosion-scarp forming this eastern shore-line, and also of nearly all the major drainagelines in the district, with the strike of the folded rocks, suggests not only that the features are subsequent, but also that they are guided by an alternation of weak and resistant beds in the highly-inclined series of grey-wackes and argillites forming the bed-rock of the district. The contrast between the weak and resistant zones appears to be due in reality to the relative freedom from joints and planes of shearing in the rocks of the latter, and to the jointed, sheared, and sometimes completely crushed condition of the former, which are perhaps best termed “shatter-belts.” These belts, however, with one notable exception—the line of the Wellington fault (Cotton, 1914)—are parallel with the strike, and appear to be the result of thrust-faulting which occurred as an accompaniment of the folding of the strata (post-Hokonui orogeny). (The positions of several fault-zones, or shatter-belts, in the Wellington Peninsula, west of the depression, were indicated by Broadgate, 1916.) Thus the ridge which bounds Port Nicholson and the Hutt Valley on the south-east, along with the straight and parallel valleys of the Mangaroa Stream (a tributary of the Hutt River) and the Wainui-o-mata and Orongorongo Rivers, and the ridges of the Rimutaka Range,

– 133 –

to the east, whether or not each is confined to the outcrop of a single formation, may be described with perfect propriety as subsequent in origin. Prior to the formation of the Port Nicholson depression it appears that the whole district was maturely dissected by the subparallel streams of the system just described, and by their numerous small insequent tributaries, with a relief of about 1,500 ft.

The erosion-scarp descending to the shore of Port Nicholson and to the Hutt Valley, as will be shown below, is not the boundary of the whole depression, this being found to be a relatively broad strip of strongly-warped land-surface.

The other boundaries of the depression call for passing reference only. On the north-western side the immaturely dissected scarp of the Wellington fault, mentioned above, meets the warped eastern slope obliquely in the Hutt Valley, which is thus a fault-angle depression. South-westward this fault-scarp extends inland about a mile and there dies out, and thence southward to the sea-coast the western side of the depression is apparently a warped surface, though evidence of the exact nature and extent of the warping has not yet come to light. Seaward, to the south, the depression is open to the Pacific Ocean.

General Tectonic Features of the Depression.

The foregoing features of the Port Nicholson depression, taken in conjunction with the observation that the Wellington fault-scarp follows a pre-existing line of weakness—a very prominent shatter-belt extending north-east and south-west, somewhat oblique to the system of subsequent features previously referred to, and marked by prominent subsequent faultline valleys (Cotton, 1914)—would seem to indicate that faulting must be regarded as merely an incident in the formation of the depression. The principal event appears to have been the sharp downwarping of a belt of land about thirty miles long elongated in a north-north-east and south-south-west direction (and extending an unknown distance farther to the south-south-west beneath the sea). The depth of downwarping that must be assumed is variable, the maximum being perhaps in the neighbourhood of 1,500 ft., or perhaps rather more, where the broadest part of Port Nicholson now is. The width of the strip affected also varies in different parts, but is at least ten miles where Port Nicholson is widest. Both depth and width diminish, though irregularly, to the north-east up the Hutt Valley.* It is as though the sagging-down in synclinal fashion of an ill-supported superficial flake of the lithosphere crossed fortuitously by the shatter-belt marking a pre-existing fault had resulted incidentally in the formation of the more modern Wellington fault, the scarp of which replaced part of the warped border of the depression. When the evidence of the features of the neighbouring coasts are taken into account, however, it appears unlikely that the harbour-depression can be accounted for so simply as by mere downward sagging owing to lack of support. The deformation of the ancient strand-lines may be ascribed to compressive folding, or the warping of the land to the east and south-east may be described as tilting of an earth-block, for, as shown below, the warping or tilting has

[Footnote] * If, as Adkin (1919) has suggested, the drowning of Porirua Harbour and the formation of Port Nicholson are due to the same movement, the downwarped strip must become wider northward, or must send out a branch towards Porirua Harbour.

[Footnote] † Similar synclinal warping with one side partly replaced by a fault occurs in the Aorere district, in northern Nelson (Cotton, 1916B).

– 134 –

taken place about a hinge-line, depression being confined to the western side of this, while to the east there is evidence of uplift only. The tilted block affected by this movement is elongated in a north-north-east and south-south-west direction, and is bounded on the east by a well-marked fault-scarp which forms the eastern front of the Rimutaka Range (Cotton, 1916, p. 318) and the fault-coast of Palliser Bay.

As will be shown below, this movement took place very recently. Such strongly differential movement of a small earth-block in very recent times is unusual even in New Zealand, though it was common enough in somewhat earlier times when the mountain masses were blocked out and the river-courses determined by the movements to which the name “Kaikoura” has been applied (Cotton, 1916). Since the Kaikoura orogenic movements took place throughout New Zealand a very great deal of erosion has occurred, but in the Port Nicholson area, on the other hand, the later stages at least of the tilting, warping, and faulting deformed and dislocated a land-surface the relief of which had already become very nearly that of the present day.

The features here described may perhaps be correctly ascribed to a modern local recrudescence of the Kaikoura movements. It is interesting to note in this connection that the latest movement which affected this area—that which accompanied the earthquake of 1855—tilted a block of considerably greater width, though bounded on the eastern side by the same fault, and that the whole district here described was uplifted, including the previously depressed harbour area (Lyell, 1868). It is as though the events which led to the formation of the harbour-depression were a belated reversion to the Kaikoura type of movement, resulting in strong local deformation of the surface, interrupting the more stately movements of larger blocks now in progress throughout the New Zealand region. The fact that the 1855 movement was of the latter type has led the writer to suspect that even in the Port Nicholson district such movements are now normal, and to formulate a working hypothesis that a succession of nearly uniform uplifts preceded the warping and tilting that formed the Port Nicholson depression. The real succession of movements has not yet been worked out with, certainty, however, and some puzzling features still remain unexplained.

It is highly probable that the warping or tilting responsible for the features here described did not go on continuously and rapidly as a single event, but was broken by pauses of considerable length. Little more than the general evidence can be considered at present, however, as it has not yet proved possible to separate satisfactorily the evidence of successive movements.

The Evidence.


The evidence of tilting and warping on the eastern side of the Port Nicholson depression is of three kinds: (1) Tilted uplifted coastal platforms, (2) progressively more extensive drowning of valleys from the hinge-line of tilting to the axis of maximum depression (which is accompanied by rejuvenation of the valleys on the other side of the hinge-line), and (3) evidence of regrading in warped valleys, particularly aggradation in such as are tilted backward.

The evidence under the first two heads is found in one line of section only, that formed by the sea-coast, while that under the third head can be seen at a number of places along the eastern boundary of the depressed area.

– 135 –

Tilted Coastal Platforms.

The Platforms farther West.—As previously noted by the writer (1912; 1916a), the coast both east and west of the entrance to Port Nicholson is bordered by remnants of platforms cut by marine abrasion when the land stood considerably lower than it does now. Around the shores of Port Nicholson itself, and on the partially drowned ridges immediately to the west of the entrance, which form part of the deeply depressed area, no traces of uplifted benches clearly referable to marine erosion are to be found above the rock platform that was raised a few feet above sea-level in 1855. Since parts of this shore-line have not suffered severe retrogradation by marine erosion, some remnants of uplifted platforms should survive if such had existed; and it may therefore be assumed from their absence that this part of the coast has always escaped uplift, or that any strands that have been uplifted have been lowered again below sea-level. The profile of the sea-bottom off shore, as revealed by soundings, does not show terraces such as would be produced by submergence of cliff-bordered remnants of cut platforms; but this negative evidence has little significance, for the initially sharp, well-defined subaqueous features that would be thus produced would soon be obliterated by deposition of the large amount of waste supplied from the neighbouring mountainous area of rather easily eroded rocks, which are subject to strong marine as well as subaerial erosion.

To the west of the depressed area the high-standing marine platforms indicate uplift and a limited amount of deformation. Two such platforms are distinctly recognizable (Cotton, 1912, fig. 7; 1916A, fig. 8), but only the lower of these can be traced for any considerable distance along the coast. At Tongue Point it is continuous as a broad shelf (except where intersected by a large ravine) for two miles, and at the ravine, where a section of the ancient beach at the rear of the platform may be seen, its height is 230 ft. In this portion no variation in the height of the rear of the platform has been observed, though its variable width, and especially its variable seaward slope, give a false appearance of irregular variation in height when it is viewed from the offing. For three miles farther westward, as far as Cape Terawhiti, the platform is traceable continuously, though it is thickly covered with talus and only at a few places is wide enough to form a distinct bench. When the cliffs are viewed from the sea, however, the top of the talus-covered wave-cut platform can be distinctly traced all the way, and where the bench is narrow this cannot be far below the ancient strand-line, which is thus seen to descend gradually to a height of about 100 ft. at Cape Terawhiti.

It appears, therefore, either that this portion of the coast was uplifted with a gentle westward tilt, or that it was uplifted more evenly and afterwards tilted gently westward; and, though the evidence cited indicates greater deformation of the ancient strand-line than the writer formerly supposed had occurred (Cotton, 1912), this deformation is slight as compared with that in the Port Nicholson area.

It may be inferred, first, that the land west of the Port Nicholson depression was uplifted with only slight deformation while Port Nicholson was sinking, the two areas being connected by a warped strip; or, secondly, that the two areas represent blocks moving quite independently (though at the coast-line there is no recognizable fault-scarp separating them); or, thirdly, that the uplift which raised the platforms on the west affected the whole district nearly evenly, and that in the Port Nicholson area the uplifted platforms have been resubmerged by more recent warping

– 136 –

—perhaps contemporaneously with the gentle tilting of the western coastal platforms. In the last case, as in the first, there must be a transitional area between the permanently uplifted and the resubmerged areas. The uplifted platforms give no information as to the nature of the warping in the transitional strip, for three-quarters of a mile eastward of Tongue Point they end, their former continuation having been cut away by modern cliff-retreat.

Platforms of the Eastern Side tilted towards the Depression.—To the eastward, though not actually bordering Port Nicholson itself, coastal platforms make their appearance not far from it, and these are strongly tilted.

As on the western side, there is only one bench that can be traced for a considerable distance with certainty; and there is a temptation to correlate it with the 230 ft. platform at Tongue Point, but such correlation is by no means certain. The ancient shore-line at the rear of this bench, which may be termed the “Baring Head platform,” as it is developed around Baring Head (Plate XXIX, fig. 2), rises from 100 ft. in Fitzroy Bay (fig. 2) to 450 ft. at the mouth of the Orongorongo River (Plate XXX, fig. 1). Its continuation in both directions beyond these points has been

Picture icon

Fig. 2.—Diagram-sketch of the southern end of the tilted area east of Port Nicholson. From left (north-west) to right (south-east) the coastal features shown are: Pencarrow Head, Lake Koangapiripiri, Lake Koangatera,-Fitzroy Bay, Baring Head, Wainui-o-mata River, Orongorongo River, Cape Turakirae.

destroyed by cliff-recession, but between them it is quite continuous except for the narrow, steep-sided valley through which the Wainui-o-mata River flows out. The distance in a direct line between the two ends of this platform remnant is two miles, and along this line a tilt of 175 ft. per mile is therefore indicated. This is, of course, very distinctly visible to the eye (fig. 2, and Plate XXX, figs. 1 and 2; see also Cotton, 1916A, fig. 19), though, as the bench follows the trend of the coast around Baring Head, the whole of it cannot be seen at once except from some distance out at sea.

The Baring Head platform is very slightly dissected. It retains its discontinuous covering (in places 30 ft. deep at the present cliff-edge) of gravel and coarse sand, both derived from the local greywacke, the sand being now almost completely weathered into sandy clay Through this veneer the solid rock projects in places, some of the projections rising 20 ft. above the ground-level as stacks, which are now crumbling to

Picture icon

Fig. 1.—The eastern shore of Port Nicholson, an erosion-scarp with steep-grade ravines, embayed by partial submergence, cliffed by marine erosion, and later in part prograded.
Fig. 2.—The Baring Head platform as seen from Cape Turakirae, 450 ft. high at the eastern (right) end, and 270 ft. high at Baring Head (on the left), broken by a single gap at the mouth of the Wainui-o-mata River.

Picture icon

Fig. 1.—The Baring Head platform between the mouths of the Wainui-o-mata and Orongoiongo Rivers, with Cape Turakirae in the distance.
Fig. 2.—The Baring Head platform, Orongorongo platform, and higher benches, as seen from the extremity of Baring Head, showing the accordance of the Baring Head platform across the gap formed by the mouth of the Wainui-o-mata River.

– 137 –

decay. These stacks are exceptional features of the high-standing coastal platforms, the smoothness of which seems at first sight remarkable when they are compared with the rugged rock platforms of the modern shoreline; but the explanation seems to be that the rocky prominences soon succumb to subaerial weathering. At the extremity of Baring Head a rocky table which evidently was cut very nearly at sea-level, being perhaps bare at low, water, now forms the outer part of the platform remnant (Plate XXXI, fig. 1), and is separated from the ancient cliff behind it by the bed of a channel 6 ft. or 8 ft. lower, floored with coarse sand containing gravel lenses. This outlying reef was evidently the outcrop of a belt of resistant rock. It and the ancient channel behind it are distinctly traceable along the platform for half a mile northward.

The small streams which drain the surface of the platform appear to be all consequent. On the eastern side of Baring Head they flow directly towards the sea, but on the north-western side the little streams from the cliff above join one which flows for a few hundred yards lengthwise (northward) along the platform before turning seaward. This is quite clearly guided by the channel between the land and the ancient outlying reef described above. The northward direction agrees with the direction of tilting of the platform. No abandoned courses or gaps in the outer reef were observed which would indicate that streams had been diverted either by capture or by tilting from former more direct courses to the sea. On the surface of the platform the streams flow in widely-opened shallow valleys in the soft veneer of gravel and sand, and they have cut notches in the bed-rock only at the cliffed edge of the platform. The above description does not apply to larger, extended streams, such as the Wainui-o-mata, which crosses the platform at grade in a deep valley, or a smaller extended stream between the latter and the Orongorongo, which has deeply notched the platform.

At Baring Head there are also two very distinct remnants of benches at heights of 80 ft. and 160 ft. respectively above the Baring Head, platform, (fig. 2). At the back and front of each there is a distinct cliff, and the covering of gravel on the highest is still quite fresh. Where observed in a prominent outcrop the gravel consists of a mixture of-large and small pebbles without any finer material, and is loosely cemented with iron oxide. These benches are drained by channels which cross them at right angles. As in the case of all uplifted platforms backed by cliffs, it is difficult to determine the exact levels of the ancient shore-lines because of the amount of talus material overlying them. For this reason, together with the fact that the remnants do not extend for more than a few hundred yards along the western side of Baring Head, it was not found practicable to decide whether they are tilted to exactly the same extent as the Baring Head platform or not, though when they are viewed from a distance the impression received is that the benches are approximately parallel.

Eastward of Baring Head, between the Wainui-o-mata and Orongorongo Rivers, the two benches last mentioned do not survive, but above the eastward continuation of the Baring Head platform (which rises here to 450 ft.) there are other well-preserved remnants at a much greater height (Plate XXX, fig. 2). The most prominent of these is the next above the Baring Head platform. It is of considerable width, is submaturely dissected, and rises to a height of 900 ft. at the eastern end. This may be termed the “Orongorongo platform.” It and the two higher remnants, which are several hundred feet higher and are nearly as well preserved (Plate XXX, fig. 2), are seen from seaward to be distinctly tilted westward, and their

– 138 –

inclination in that direction appears to be parallel to that of the Barine Head platform below them.

None of these platforms can be traced with certainty beyond the Orongorongo River and around Cape Turakirae, which forms the end of a narrow mountain-ridge, but the profile of the ridge is smooth towards the end, and there is a faint suggestion of a bench rounded and lowered at the margin by erosion, the rear of which is at a height of about 1,200 ft. This may perhaps be the continuation of the Orongorongo platform (Plate XXX, fig. 1).

Westward also the continuation of these higher platforms is doubtful, but the Orongorongo bench may perhaps be correctly correlated with a platform of considerable extent forming the crests of the ridge and spurs between the Wainui-o-mata valley and the lake (Koangatera) at the mouth of the Gollan's Valley stream. This platform, which may be named the “Wainui platform,” is submaturely dissected, and is 500 ft. high at its rear on the ridge west of the Wainui-o-mata valley. It slopes southward (towards the sea) fairly steeply, but not so steeply that its slope in that direction cannot be explained as probably original.

The points at which the observations of the height of the rear of the Orongorongo platform as 900 ft. and of the Wainui platform as 500 ft. were made are two and a half miles apart, in a north-westerly direction, and thus, on the assumption that these two may be correlated, the tilt indicated is the same as that of the Baring Head platform.

Farther to the west the Wainui platform, still descending, is cut through by Gollan's Valley, on the eastern side of which the height of the rear of the platform is 340 ft. and on the western side only 180 ft. Here (between the lakes Koangatera and Koangapiripiri) the platform is half a mile wide and very distinct, though submaturely dissected. It is not shown on the published contoured map of the district, which at this point is not quite accurate. There is no trace of this platform, or of any others, farther on around or beyond Pencarrow Head.

Besides these remnants of uplifted and tilted coastal plains there is eastward of Baring Head a less strongly uplifted strip of recently emerged sea-bottom, which extends round Cape Turakirae (fig. 2, and Plate XXX, fig. 1) and along the western shore of Palliser Bay (Aston, 1912). It will be referred to as the “Turakirae coastal plain.” The greater part of this coastal plain, though its seaward slope is very steep (about 1 in 10), is not yet cliffed at the margin. Strangely enough, it was not found to be tilted to the westward, as the higher benches are. The absence of evidence of tilting, and the difficulty which this raises as to the non-continuation of a feature indicating such recent uplift along the coast westward of Baring Head, necessitate the introduction of a brief description of the Turakirae coastal plain; but, since this would make too long a digression at the present stage of the presentation of the evidence of tilting, it is placed in an appendix.

Evidence from Drowned Valleys.

The tilting of the block east of Port Nicholson on a hinge-line, which may be assumed in explanation of the tilted uplifted platforms described above, involves partial or complete submergence of its north-western edge, and this is found not only in the drowning of the central part of the Port Nicholson depression to form Port Nicholson itself, but also, nearer at hand, in the drowned mouths of two small valleys opening between Pencarrow Head and Baring Head. As the shore-line is followed westward from Baring Head towards Port Nicholson for some distance the mouths

– 139 –

of small ravines only are passed, and these hang above the shore-line as a result of cliff-recession that has recently been in progress. Slight submergence produces no noticeable effect on the mouths of hanging valleys such as these, and so the exact position of the hinge-line of tilting cannot be ascertained from them. Farther on, however, Gollan's Valley and the valley of a small stream debouching close to Pencarrow Head are drowned to such an extent as to indicate very considerable submergence (fig. 2, and Plate XXXI, fig. 2).

Gollan's Valley is fairly large, heading eight miles inland, but the other is only two miles long. The streams in both are of such size, however, that it may be supposed that they reached the sea at grade prior to submergence. It is clear that both, when first drowned, were occupied by winding lanes of sea-water, the one three miles and the other rather more than a mile in length. These bays are now cut off from the sea and converted into fresh-water lakes by gravel bars 20 ft. in height above sea-level and accordant with the pre-1855 storm-beach ridge, which is well developed along the neighbouring shore-line in Fitzroy Bay. At the western end of each bar the outflowing water has opened a channel through the gravel.

The enclosed bays are much reduced in size by the growth of swampy deltas at their heads, and the length of the two lakes which now occupy their lower parts (Koangatera in Gollan's Valley and Koangapiripiri in the other) is thus reduced to about half a mile in each case. The upper part of Gollan's Valley is also thickly aggraded with alluvium. Near the mouths the sides of both valleys are cliffed, and out-jutting points are strongly truncated (Plate XXXI, fig. 2). The cliffs reach a height of 100 ft. on the shores of Koangatera and 50 ft. around Koangapiripiri, and they are evidently the work of waves at a time when the bays were still deep and open to the ocean. For a mile up the somewhat winding Gollan's Valley the swampy delta is bordered, however, by low wave-cut cliffs. These must be the work of waves raised on the narrow landlocked waters, and their presence indicates a long period of still-stand prior to 1855, for the relative levels of sea and land were constant long enough not only for the development of distinct cliffs (though on mature hill-slopes of weathered rocks, it is true) by waves with a fetch of no more than a few hundred yards, but also for the delta-front to advance for quite a mile past the farthest inland point where cliffs are traceable. This indicates that the uplift of 1855 was either the precursor of a new series of earth-movements or was an isolated phenomenon; and the shore-line features at the western side of the entrance to Port Nicholson lend support to this view.

As the writer has shown elsewhere (Cotton, 1921), this is a matter of great practical importance. If movements like that of 1855 had been common in the immediate past the outlook for the future safety of the city of Wellington and the continued usefulness of its harbour would be rather poor; for it must be remembered that the cause of the disastrous earthquake of 1855 was directly connected with the uplift which then occurred, and also that the harbour was made fully 5 ft. shallower by the same movement. As it is clear from what has been stated above that the 1855 movement was the first of its kind in this district for thousands of years, the outlook for Wellington is distinctly hopeful.

A rough indication of the measure of the submergence shown by the drowning of the valleys to form the bays now occupied by the lakes Koangatera and Koangapiripiri can be obtained by comparing the widths of the mouths of the embayments with the widths of similar but unsubmerged valleys in the district at various heights above the floor, and also

– 140 –

by comparison of the lengths of the drowned portions with the average gradients of similar valleys. For this purpose the contoured map of the district has been used. The result shows that the depth of submergence at Gollan's Valley may be as much as 200 ft., but cannot be more. The Koangapiripiri valley is so small that the result obtained by comparison with the small-scale map available is not very reliable, but the submergence indicated does not seem to be greater than in Gollan's Valley, though indications of greater submergence might be expected, seeing that it is halfway between the latter and the deeply-drowned entrance to Port Nicholson.

The evidence of the drowned valleys supports the hypothesis of fairly even tilting towards the Port Nicholson depression. The position of the hinge-line is not definitely indicated. It is possible that the mouth of the Wainui-o-mata valley has been very slightly drowned and quickly filled again with alluvium by the river, but this is doubtful. Farther east, however, at the mouth of the Orongorongo, there has been no submergence, for this river is cutting on bed-rock at the mouth.

Some of the rejuvenation of which there is evidence in the Orongorongo valley is in all probability a result of the uplift of the eastern side of the tilted area which caused the emergence of the Turakirae coastal plain. Hence the rejuvenation of the eastern valleys is not such good proof of tilting as the drowning of the western valleys.

Evidence from Regraded River-valleys.

Near the coast the larger streams flow without exception in courses parallel with the hinge-line of tilting inferred from the evidence already described. They are close together, and have only very small, steepgraded tributaries. In this part of the district there are, therefore, no streams that would be particularly sensitive to tilting in a west-north-westerly direction towards the Port Nicholson depression. Degradation in the Orongorongo and Wainui-o-mata valleys, of which there is evidence in the presence of low terraces, might be the result of purely regional movement, and distinct terraces can be correlated with uplifted strand-lines on the Turakirae coastal plain, which lies across their mouths. The aggradation in Gollan's Valley, which has previously been described, can be accounted for by the drowning of the valley-mouth. Nevertheless the western branch of Gollan's Valley is aggraded quite to the head, as though as a result of headward tilting, which would be the result of the general downwarping of the surface towards Port Nicholson.

There is a suggestion of aggradation also in some parts of the valley of the Wainui-o-mata, which is somewhat winding. It is a mature valley, with a flood-plain, and this widens out considerably in places, where the valley bends to the west. Though such expansions are due in part to the development of large curves by lateral planation and their later abandonment when the stream returned by a cut-off to a straighter course, they appear to be partly the result also of aggradation in response to backward tilting in certain reaches, either because these are below westward bends in the sinuous valley, or else on account of transverse waiping, such as certainly occurs farther north, corrugating the general slope of the country towards the Port Nicholson depression.

These aggradational effects are, however, much less definite proof of tilting than some which are found farther north, opposite the Hutt delta and the upper Hutt Valley. Here, though the principal valleys still trend north-north-east or south-south-west (Cotton, 1914), a number of tributaries of moderate size enter them from the west. The western branch of the

Picture icon

Fig. 1.—The Baring Head platform on Baring Head, showing a flat-topped ancient outlying reef surmounted by stacks.
Fig. 2.—Lake Koangatera, at the drowned mouth of Gollan's Valley, showing strongly-cut sea-cliffs along the sides and bordering the bay-head delta in the foreground.

Picture icon

Fig. 1.—The aggraded, headward-tilted valley of the western branch of the Wainui-o-mata, as seen from a point on the divide at its head.
Fig. 2.—View looking seaward across the widest part of the rocky coastal plain of Cape Turakirae, showing successive beach-ridges build during pauses in the uplift.

Picture icon

Fig. 1.—Map of the Hutt River delta, which partly fills the Port Nicholson depression, and the aggraded valaley of the western branch of the Wainui-o-mata. Approximate scale; ½ in. + = 1 mile. (From the ½ in. topograhic map.)
Fig. 2.—Map showing the Upper Hutt basin-plain (an expansion of the north-eastern continuation of the Port Nicholson depression) and the aggradel tributaries of the Mangaroa. Approximate scale; ¼ in. = 1 mile +. (From the map of the Trentham Manoeuvre Area.)

Picture icon

An aggraded plain in small headward-tilted tributaries of the Mangaroa River. In the centre the fans are confluent across the neck of a spur, the continuation of which emerges from the alluvium in the left middle distance. Angle of view, south to west.

– 141 –

Wainui—o-mata (Plate XXXII, fig. 1) is a good example of a backward-tilted valley. It is occupied by an aggraded plain almost to the divide at its head, as shown by the contoured map (Plate XXXIII, fig. 1), but the aggradation does not extend far down the main valley beyond the junction, where there is merely a flood-plain in the valley-bottom. Another branch of the Wainui-o-mata, Moore's Valley, is aggraded also, but not to its head, for this does not turn so far westward.

Very striking topographic features resulting from aggradation which must be the result of headward tilting occur in a tributary entering from the west the Mangaroa River, itself a tributary of the Hutt (Plate XXXIII, fig. 2). The Mangaroa is one of the apparently subsequent, north-north-eastward-flowing streams, and so is approximately parallel to the hinge-line of the most prominent tilting. The tributary referred to, or, more correctly, the several small streams which join to form it, flow at right angles to this direction—i.e., east-south-east, or directly against the slope of the land-surface which a little farther on descends below the alluvium in the Hutt Valley (a north-easterly extension of the Port Nicholson depression). It is clear that, before they were tilted headward by an earth-movement, these small streams flowed in courses approximately the same as those they now follow across a maturely dissected surface not very different from that now existing, and the present topographic features indicate that the tilting was so sharp that these steep headwater streams, with declivities as steep as several hundred feet per mile, were caused to aggrade vigorously, so as to spread an extensive sheet of alluvium. (At some stage they may have been ponded, though any evidence of such ponding is now buried beneath alluvium.) The valley-floors are now broad fan-like slopes of alluvium, the greater part of which lies just below the 800 ft. contour (Plate XXXIII, fig. 2). The fans are steep at the valley-heads, where they extend almost to the divide, and lower down are swampy except where they have been drained artificially. They spread out widely in the middle parts of the tilted valleys and taper away towards the junction with the Mangaroa Stream. To such an extent are the valleys filled about the middle of their slope that the fans in two of them have become confluent across the neck of a spur, the end of which now stands as an island surrounded by alluvium (Plate XXXIII, fig. 2, and Plate XXXII, fig. 2).

A few miles north-eastward, at Wallaceville, the Mangaroa Valley is itself aggraded. Some aggradation is to be expected as the axis of the Hutt Valley is approached. for the latter, as mentioned above, is in some places deeply filled with alluvium, and from it aggraded plains extend some little way up tributary valleys, converting them into embayments. Such aggradation does not, however, extend far up small valleys that are strongly tilted down-stream. Above the partial filling due to alluviation outside their debouchures streams entering the depression from the slope of the tilted surface ought, in general, to be rejuvenated. In spite of its belonging to this class of streams the valley of the Mangaroa is aggraded for several miles; for the lower part of the valley, which crosses the tilted block-surface very obliquely, appears to have suffered headward tilting owing to its crossing the transverse corrugation which causes the Hutt Valley to expand so as to become a basin-plain at Trentham and Upper Hutt (Cotton, 1914). This introduces into the tilting for some distance a south-westward component, which appears to be the cause of the aggradation in the Mangaroa, and more especially in Black Creek, a small tributary coming in from the south-west. The valley of the latter is filled in to form a swampy plain a mile wide and three miles in length, which shows up very conspicuously on the contoured

– 142 –

map (Plate XXXIII, fig. 2). This plain (the Mangaroa Swamp) is so level that it is highly probable it is in part a filled-in lake due to warping, rather than a plain wholly formed by aggradation.

The Mangaroa River near its junction with the Hutt is now degrading again, and this rejuvenation is shared by the upper part of the Hutt River and its other tributaries. It is due apparently to steepening of the upper course of the Hutt by the latest warping movements.


The Problem of the Turakirae Coastal Plain.

As mentioned previously, there is, to the east of the Port Nicholson depression, besides the uplifted and tilted platforms that have been described a less-strongly uplifted coastal plain, which fringes the coast eastward of Baring Head, around Cape Turakirae, and along the western shore of Palliser Bay. This feature has been described by Aston (1912) as “the raised beaches of Cape Turakirae.” Though, on account of its roughness, “plain” may not seem an appropriate term to apply to it, the fact that it is a recently-emerged strip of sea-bottom brings it into the class of coastal plains as defined in systematic geomorphology. As the deposit of gravel and boulders on it is merely a thin and discontinuous veneer, and the obvious stacks and many of what are apparently boulders remain attached to bedrock, it might also be described as a plain of marine erosion. Aston referred to part of it as a “boulder plain.”

Though the seaward slope of the coastal plain is very steep (about 1 in 10), the greater part of it is not yet cliffed at the margin. Unlike the higher benches, the Turakirae coastal plain is not tilted to the westward—or, at any rate, is not tilted to an appreciable extent. Aston found the height of the highest strand-line, at the rear of the plain, to be 95 ft., while the width varies from 250 to 400 yards.

The great size of the boulders and the general ruggedness of the former sea-bottom correspond with the exceptional steepness of its profile, which allowed the sea to abrade the cut platform and attack the cliffs behind the former shore-line with the energy of the ocean-waves practically undiminished by the friction of the bottom.

The plain is widest at Cape Turakirae, and it tapers off and ends about five miles north-eastward.

In addition to that developed prominently in places at the rear of the plain (No. 5 beach of Aston), Aston has recorded the presence of four other storm-beaches at lower levels, built during pauses in uplift. Those which he terms Nos. 4, 3, and 2 are at heights of 80 ft., 60 ft., and 40 ft. above sea-level. The latter two are very prominent and continuous around Cape Turakirae (Plate XXXII, fig. 2). The beach termed No. 1 by Aston he regarded as formed prior to 1855 and uplifted by the movement accompanying the earthquake of that year. Recent observations by the writer show, however, that parts of this beach have now been reworked and incorporated into the modern beach, as has occurred also on some other parts of the coast near Wellington. In some places, no doubt, there was no storm-beach ridge prior to 1855, while in other places the pre-1855 ridge has since been destroyed by marine erosion.

The continuous raised beach-ridges ought to record any tilting alongshore that has occurred since their formation. Down-tilting following even uplift would cause the raised beaches, when traced laterally, to disappear successively beneath the present sea-margin, while progressive tilting during

– 143 –

uplift would be indicated by convergence of the beach-ridges. While aneroid observations do not indicate tilting (Aston, 1912),-spirit-levelling of the 40 ft. and 60 ft. beaches might show a small amount of the latter kind. The higher beaches are largely smothered by fans, and correlation of isolated parts is somewhat uncertain. So these would be less suitable for testing by precise levelling.

The writer's observations, though his investigation of the Turakirae coastal plain has been very incomplete, have convinced him that the tapering-away of the plain at each end is due not to tilting, but to the rocky platform being cut away by marine erosion as the land rose. At the north-eastern end the beaches are smothered by fans, which must conceal also the cliffed margin of the higher part of the plain. The lower part of the plain continues some distance farther as a narrow bench 20 ft. above the sea, with a cliffed margin.

Towards the western end of the plain the higher beaches are smothered in a similar way by fans, but beyond these there is at the western side of the mouth of the Wainui-o-mata a small cliffed remnant of a beach-covered bench which appears to be at exactly the height of the rear of the plain at Cape Turakirae. The platform which bears the 60 ft. beach-ridge is also distinct on both sides of the mouth of the Wainui-o-mata; but farther west there is no further trace of the coastal plain, though in Fitzroy Bay there is a modern storm-beach ridge 10 ft. to 12 ft. above mean sea-level, and also a raised beach-ridge, presumably the pre-1855 beach, 10 ft. higher.

The discontinuity of the Turakirae coastal plain remains an unsolved problem. There is no trace of it around the comparatively sheltered inner shores of Port Nicholson, or, indeed, anywhere west of Baring Head. If it has been exposed by differential uplift, or its continuation resubmerged by differential subsidence, no fault-scarp boundary between the differently-moving blocks has been traced, and if this boundary is a warped surface the warped part of the Turakirae coastal plain must have been cut away by marine erosion. It is clear, however, that the even uplift of the Turakirae coastal plain took place later than the warping described in the body of this paper.

List of Papers referred to.

Adkin, G. L., 1919. Further Notes on the Horowhenua Coastal Plain and the Associated Physiographic Features. Trans. N.Z. Inst., vol. 51, pp. 108–18.

Aston, B. C., 1912. The Raised Beaches of Cape Turakirae, Trans. N.Z. Inst., vol. 44, pp. 208–13.

Bell, J. M., 1910. The Physiography of Wellington Harbour, Trans. N.Z. Inst., vol. 42, pp. 534–40.

Broadgate, F. K., 1916. The “Red Rocks” and Associated Beds of Wellington Peninsula, Trans. N.Z. Inst., vol. 48, pp. 76–86.

Cotton, C. A., 1912. Notes on Wellington Physiography, Trans. N.Z. Inst., vol. 44, pp. 245–65.

—– 1914. Supplementary Notes on Wellington Physiography, Trans. N.Z. Inst., vol. 46, pp. 294–98.

—– 1916. The Structure and Later Geological History of New Zealand, Geol. Mag., dec. vi, vol. 3, pp. 243–49, 314–20.

—– 1916A. Fault Coasts-in New Zealand, Geogr. Rev., vol. 1, pp. 20–47.

—– 1916B. Block Mountains and a “Fossil” Denudation Plain in Northern Nelson, Trans. N.Z. Inst., vol. 48, pp. 59–75.

—– 1918. The Outline of New Zealand, Geogr. Rev., vol. 6, pp. 320–40.

—– 1921. For how long will Wellington escape Destruction by Earthquake? N.Z. Journ. Sci. & Tech., vol. 3, pp. 229–31.

Lyell, C., 1868. Principles of Geology, 10th ed., vol. 2, p. 82.