Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 48, 1915
This text is also available in PDF
(2 MB) Opens in new window
– 59 –

Art. VI—Block Mountains and a “Fossil” Denudation Plain in Northern Nelson.

[Read before the Wellington Philosophical Society, 27th October, 1915.]

Plates IV, V


Introduction 59
“Block” features throughout New Zealand 59
Northern Nelson 60
The Aorere-Gouland depression 62
  The Wakamarama fault-scarp 64
 The north-eastern portion of the scarp 65
Tilted and stripped plateau of the south-eastern side of the Aorere Valley 66
Erosional, sedimentary, and deformational history 68
Dissection 69
The actual valley of the Lower Aorere 69
The Gouland Downs depression 70
  The floor 70
  The eastern boundary 72
A “catenary” saddle 73
The Slate Range 74
The drainage and dissection of the Gouland Downs 74


For some years the writer has been interested in the geomorphogeny of northern Nelson, and in 1913, though at that time without personal knowledge of the district, he presented a brief note on the subject before the Geological Section of the Wellington Philosophical Society, questioning in some measure the interpretation of the relief given by Bell in the Parapara bulletin of the Geological Survey.* The views then expressed having been favourably received by geologists acquainted with the district, and Professor W. M. Davis having in the meantime advised him to take up the subject of block mountains in New Zealand, the writer paid two visits to northern Nelson in 1915, and is now able to give a detailed description of a small portion of the district, and to express some general opinions as to the remainder. In this paper a condensed and generalized description of northern Nelson is given, and that is followed by a more detailed description of some features of the Aorere Valley and of the Gouland Downs.

“Block” Features throughout New Zealand.

As a result of observations made in many parts of New Zealand at various times, and confirmed by special visits recently made to a number of critical localities, the writer has come to the conclusion that the present relief is very largely—almost entirely—due to recent differential movement of crust blocks, both large and small. Though this explanation of the relief is not to be found in any general work on the geology

[Footnote] * J. M. Bell, “The Geology of the Parapara Subdivision,” N.Z. Geol. Surv., Bull. 3, 1907.

– 60 –

or geography of New Zealand, it is by no means altogether new. It was clearly in the mind of McKay as early as 1883,* and it is implied in the descriptions of various areas in Nelson and Westland made during the last decade by Henderson, Morgan, Webb, and others.

The structure of the younger rock formations, where these have been preserved, affords ample confirmation of this explanation of the relief Most of the larger relief features are tectonic forms—of course modified by erosion to a greater or less extent—while the river-courses are very largely consequent, still following very closely the courses taken upon the tumbled and irregular surface produced by a late disorderly uplift.

In many cases the blocks or units of the disorderly tumbled crust are bounded by faults. This is, however, by no means universally true, and where either formerly horizontal strata—that is to say, horizontally deposited strata which are so young that they must be supposed to have lain in their original horizontal attitude in the period immediately preceding that of uplift—or formerly horizontal, planed surfaces are present, these frequently exhibit evidence of considerable deformation by folding. In some parts of the country, indeed, the intense folding and mashing that the young strata have undergone point to strong compression accompanying—perhaps initiating—the movements. In other parts, however, such evidence is lacking, formerly horizontal strata and planed surfaces remaining flat, though frequently tilted, over considerable areas. In the present state of our knowledge it cannot be stated whether normal or reverse faults predominate. No attempt will be made in this paper to explain the cause of any earth-movements, but attention will be directed to their effects as seen in the present form of the land-surface.

New Zealand may be described as a concourse of earth-blocks of varying size and shape, in places compressed, the highest blocks lying in the north-east and south-west axis of the land-mass, so that the whole structure may be termed a geanticline, the blocks initially consisting of an undermass of generally complex structure much denuded and largely planed, and covered over most of the area by an overmass which had not been disturbed before the “blocking” movements took place; the whole since the movements considerably modified by both degradation and aggradation.

In the present paper, a sketch of but a small portion of the New Zealand area, the writer cannot hope to demonstrate the accuracy of the foregoing general description, but it is offered as a working hypothesis which gives much assistance in the interpretation of this particular district.

Northern Nelson

In the north-western corner of the South Island two fault-angle depressions, the Aorere and Takaka Valleys (see Fig. 1), open out broadly towards the north-east and north, separating three composite upland blocks which, owing to diminishing throw of the boundary faults and consequent dwindling of the fault-angle depressions towards the south-west, coalesce in that direction. The north-western, or Wakamarama, block presents a fault-scarp front, but little dissected, towards the Aoreie Valley, while north-westward its back slope, much dissected by consequent

[Footnote] * A McKay, “On the Origin of the Old Lake Basins of Central Otago,” Geol. Surv of N.Z, Rep Geol Expl dur 1883–84. pp. 7681, 1884; see also “On the Geology of Marlborough and South-east Nelson,” ibid, 1890–91, pp. 1–28, 1892.

– 61 –

streams, descends towards the Tasman Sea. The mountain-ranges of the middle, or Haupiri, block—that between the Aorere and Takaka Valleys—appear to have been carved from a mass which had initially a rough anticlinal or domed form, its present surface descending towards the north-west, north-east, and east from heights of over 5,000 ft. at the south-western end. The block is perhaps composed throughout of a number of smaller or secondary blocks separated from one another by faults and flexures. This is certainly the case towards the west, where the Haupiri block coalesces with the Wakamarama block in a region of flexed and broken plateaux known as the Gouland Downs. These plateau surfaces are remants of a stripped, “fossil” denudation plain

Picture icon

Fig. 1.—Locality map of northern Nelson.

which formed the floor underlying a series of marine strata since removed over large areas, but here and there preserved.

These marine covering strata are weak in comparison with the resistant undermass, and so the planed surface of the latter, which truncates indifferently indurated and metamorphosed clastic sediments and crystalline limestone of complex structure and also granitic intrusions, and forms the floor upon which the covering strata have lain, survives in favourable situations as a plateau long after the cover has been removed. It has, however, been destroyed, as might be expected, where its attitude favours deep dissection.

The Takaka Valley fault-angle depression is bounded on the east for twelve miles by an almost undissected fault-scarp nearly 3,000 ft. in height, with a north-and-south trend, which is the western edge of the block forming the Pikikiruna Range, and which may, therefore, be

– 62 –

appropriately named the Pikikiruna fault-scarp (see Fig. 2). The Pikikiruna block, which is tilted towards the east, is much dissected on its eastern or back slope, the recently drowned margin of which forms the indented western shore of Tasman Bay.

Picture icon

Fig. 2.—View looking south along the Pikikiruna fault-scarp, which bounds the Takaka Valley on the east. In the centre is seen a hog-back of Tertiary limestone, which is turned up along the fault.

Tasman Bay itself no doubt had its origin in the subsidence of an earth-block, while the lowlands forming its southerly continuation have the appearance of a fault-angle depression, now much modified by erosion, bounded on the east by the scarp of the Richmond or Waimea fault* and on the west by the eastern boundary, probably in great part a dissected back slope, of the great block or complex of blocks constituting the highlands of Mount Arthur, the Mount Arthur tableland, and the neighbouring ranges, which are more or less continuous with the Haupiri and Pikikiruna blocks towards the north. The structure of the covering strata as interpreted in a series of sections by McKay, however, indicates a considerable complication of the block movements in this neighbourhood by folding.

The Aorere-Gouland Depression.

The Aorere River in the lower seventeen miles of its course, in which it flows north-eastward, is guided by the fault-angle depression previously referred to as the Aorere Valley, a name which it is convenient to restrict arbitrarily to this obviously consequent portion of the whole river-valley. At the head, or south-western end, of the Aorere Valley the river enters it from the south, emerging from a deep, narrow, and steep-sided valley between high mountains—a valley which, unlike the other, appears to owe the whole of its depth and width to erosion, which is perhaps consequent but possibly insequent, and which may be conveniently designated the “Upper Aorere Valley”.

The Aorere Valley depression is bounded on the north-western side by the fault-scarp front of the Wakamarama block, and on the south-eastern side by the tilted surface of a portion of the Haupiri block. To the north-east it is open to the sea, while at the south-western end,

[Footnote] * See Bell, Clarke, and Marshall, “The Geology of the Dun Mounta Subdivision, Nelson.” N Z Geol. Surv, Bull No 12. p. 12, 1911.

[Footnote] † A. McKay, “The Baton River and Wangapeka Districts and Mount Arthur Range,” Geol Surv of N Z, Reports of Geol Expl dur 1878–79, p. 122, 1879.

– 63 –

just beyond the point at which the Upper Aorere Valley opens to the south, the Aorere Valley is terminated by a maturely dissected scarp, certainly of tectonic origin and probably a fault-scarp, the streams dissecting which supply the Aorere with a tributary of considerable size, Brown's River. Though this scarp forms the boundary of the Aorere Valley as here defined, and though its crest is a divide between the streams of the Aorere system and those flowing westward to the Tasman Sea, it does not terminate the tectonic depression of which the Aorere Valley forms a part, for, beyond the scarp, which constitutes a step upwards of 2,000 ft. in height, the depression is continued in a south-westerly direction towards the western coast, the floor of this portion being the plateau known as the Gouland Downs. The composite depression as a whole may, therefore, be conveniently termed the Aorere-Gouland depression.

Picture icon

Fig. 3.—Locality map of the Aorere-Gouland area.

The continuation of the Aorere Valley depression in the Gouland Downs was recognized by Bell, who regarded the depression as an ancient strait in which the marine Tertiary strata found in the Aorere Valley and on the Gouland Downs were deposited. Bell's explanation will be referred to again on a later page.

The geological evidence as to the fault-angle origin of the Aorere Valley has been clearly stated by McKay* (p. 10), who summarizes his views in the following words: “The disposition of the rocks in every

[Footnote] * A. McKay, “The Geology of the Aorere Valley, Collingwood County, Nelson,” Papeis and Reports rel. to Minerals and Mining, Parl. Paper, C.–11, 1896, pp. 427.

– 64 –

part of the Aorere Valley indicates that it was first formed along a line of earth-fracture, trending in the general direction of the present valley, and having its downthrow on the south-east side of the line of rupture.” (p. 9).

The Wakamarama Fault-scarp.

The scarp front of the Wakamarama Range, which forms the north-western side of the Aorere Valley depression, belongs to a class of composite fault-line- and fault-scarps common in New Zealand, its upper part being a true fault-scarp, and its lower part, from which the covering strata of the downthrown block have been in part stripped by erosion since faulting took place, being a fault-line scarp. This “composite fault-scarp,” as it may be called, rises to a height of about 3,000 ft. above the valley lowland. Only one stream—the Kaituna—breaks through the scarp and has its head far back in the range, and, with the exception of this break, the divide between the heads of the numerous steep-grade consequent ravines of the scarp and the north-westward-flowing rivers of the back slope of the Wakamarama block lies at no great distance back from the fault-line. The edges of the facets into which the scarp is divided by the iavines which dissect it are rounded off, and no doubt the steepness of the facets has been much reduced by weathering, slipping, and soil-creep. In spite of a covering of forest which obscures the details, however, the alignment of the blunt-ended spurs is still very striking; and, even in the absence of the geological evidence of the existence of a fault along this line, which is afforded by the presence of the covering strata in the valley lowlands along the base of the scarp, the morphological evidence would indicate faulting. In that case it would be necessary to give very careful consideration to an alternate hypothesis that the scarp is the result of lateral cutting in the course of normal erosion by the Aorere River, which flows at its base, or perhaps that it has been formed by glacial erosion. The most convincing argument against either of these hypotheses is the absence of a similar scarp on the opposite side of the Aorere Valley. Though the whole scarp cannot by any stretch of the imagination be regarded as the work of fluviatile erosion, the streams of the Aorere system, while they were engaged in eroding away the initial floor of the depression, undoubtedly swung occasionally against the base of the scarp, which may thus be expected to be now a little way back from the fault-line.

For the greater part of the length of the scarp the undermass rocks alone occur in the upthrown block in the neighbourhood of the fault-line, and the crest appeals from the valley as a line of rounded prominences of roughly accordant height, while, according to Bell, there are on the highland surface patches of pakihi (flat open spaces) (p. 24), suggesting that the range is a dissected block of an uplifted denudation plain. From the crest of the range, which, as stated above, is but a short distance back from the fault-line, the summit-levels descend towards the north-west; and the north-western slope is dissected by large sub-parallel streams of apparently consequent (more strictly, probably, superposed consequent) origin. The lower ground to the north-west is formed of a continuous sheet of the covering strata, maturely dissected, and there are outhers of the same towaids the crest of the range* All

[Footnote] * Cox, S. H., “On the District between Collingwood and Big River,” Geol Surv. of N.Z., Rep. Geol. Expl. dur. 1882. pp. 6274, 1883, Park, J., “On the Geology of Collingwood County. Nelson.” ibid., 1888–89. pp. 186243, 1890; Bell, J. M., loc cit, map opp. p. 89

– 65 –

these facts give strong support to the view that the Wakamarama Range is the dissected remnant of a tilted block (or possibly a complex of minor folded and faulted blocks which may in a general description be conveniently considered as a single block) presenting its front or scarp to the Aorere Valley and its back slope to the Tasman Sea, a view involving the not unreasonable assumption that the covering strata, lying on a planed surface of the undermass, were continuous across the site of the range prior to the uplift of the block, and that since the uplift they have been more completely removed from the higher than from the lower ground.

As the foregoing assumption is contrary to the interpretation of the geological history and physiography of the district given by Bell in the Parapara bulletin to which reference has already been made, the conclusions there arrived at may be also stated. According to Bell's interpretation, maturely dissected mountains occupied the area in the period immediately preceding that in which the covering strata were laid down, and the period of deposition was one of only partial submergence. The unsubmerged mountains are regarded as still surviving in a form but little altered, and, on account of their supposed relation to the younger deposits, they are termed the “old land.” While the occurrence of some faulting and therefore of some differential movements of later date than the period of submergence is recognized, the uplift of the already mountainous “old land” to its present height is ascribed mainly to “bodily secular movement since the Miocene era” (pp. 21, 2324).

“The old land represents physiographically an ancient mountain-lange which had probably been maturely dissected prior to Miocene times. One sees generally the rounded outlines so characteristic of elevated land-surfaces subjected to long-continued subaerial erosion.” (p. 23).

The North-eastern Portion of the Scarp.

Towards the north-east, as on the north and north-west, the surface of the Wakamarama block descends towards the sea, and maturely dissected covering strata survive on it. The fault-scarp facing the Aorere Valley decreases in height towards the north-east, therefore, and towards the mouth of the Aorere Valley the covering strata make their appearance on the crest of the range. The line of unconformable contact between the undermass and the cover runs obliquely up the face of the fault-scarp towards the south-west, but, unfortunately, owing to the covering of forest, details of the contact are not easily seen. At first sight it appears as though the even crest of the range on the undermass is continued on the bevelled edges of the cover as in Fig. 4, a. As there can be little doubt that the even crest on the undermass is determined by the resurrection of a denudation plain, this would mean the presence of intersecting denudation plains, and would involve strong unconformity between a lower series of conglomerates and coals* (those exposed on the face of the scarp) and an upper series (also coal-bearing)* on the back slope of the Wakamarama Range farther to the south-west. As, however, the beds are all regarded by Park as belonging to one conformable series, it is probable that the appearance of intersecting denudation plains is false, arising from an increasing inclination

[Footnote] * Described by Cok and Park in the papers previously referred to.

[Footnote] † Bell, loc. cit., p. 55.

– 66 –

towards the north-east of the covering strata and of the floor on which they lie, as shown in Fig. 4, b, combined with resistance to erosion offered by indurated conglomerate at the base of the cover.

Tilted and Stripped Plateau of the South-eastern Side of the Aorere Valley

Turning now to the south-eastern side of the Aorere Valley, one's attention is arrested by a tilted denudation plain which rises from the valley lowland to a height of about 1,500 ft in a distance of three miles and a half. This plain from some points of view appears very well preserved, but closer inspection reveals the fact that it is deeply divided by the steep-walled gorges of numerous streams tributary to the Aorere. The form of this south-eastern slope was noted by Hochstetter,* who estimated its inclination as 8°, but did not remark upon its genesis. It was remarked upon later by Park (p. 198), McKay (1896, p. 23), and Bell (pp. 2425), McKay explaining it as a plain of marine denudation and Bell referring to it as an “old sea-shelf.”

Picture icon

Fig. 4.—Alternative interpretations of the Wakamarama fault-scarp section.
Fig. 5.—Dissection of a sloping surface, such as that of the south-eastern side of the Aorere Valley, by streams from a higher block behind it. The initial form is shown in the right-hand block.

McKay describes it as a “feature characterizing the east side of the Aorere Valley,” and adds, “This is an uniform slope of the country to the north-west and the low grounds of the valley, from heights 1,200 ft. to 1,500 ft. above the sea, which slope, as seen from a distance, appears to be remarkably uniform, both as regards its dip towards the low grounds and as regards its extension along this side of the Aorere Valley, and suggests at once the idea of a plane [sic] of marine denudation, which, by the elevation of the mountain region to the south, has acquired a steeper slope than it had when first formed.” (1896, p. 23). Illustrations are given by Hochstetter (woodcut on p. 103) and by Bell (pl. i, lower view, and pl ii, upper view).

[Footnote] * F. v. Hochstetter, “New Zealand,” Stuttgart, 1867, p. 105.

– 67 –

This sloping plateau is certainly, as McKay noted, a plain of marine denudation—a plain of denudation, that is to say, to the formation of which the finishing touches at least were given by marine erosion as the sea advanced over it and deposited upon it the covering beds since largely removed. The occurrence, however, of terrestrial formations at the base of the covering strata in neighbouring areas, notably on the north-western and northern parts of the Wakamarama Range, is perhaps an indication that the surface of the undermass had there been reduced to small relief by subaerial agencies in the period preceding submergence. It is, therefore, reasonable to suppose that here also a peneplain was in existence, and that the plain of marine erosion is one of those produced by the stipping and removal of the waste mantle from a peneplain, accompanied by a minimum of erosion of the fresh underlying rock. The large extent of the planed surface—which, indeed, probably extended formerly over a much larger area than that in which the plain is now preserved—thus receives a simple and probably correct explanation.

Picture icon

Fig. 6.—View looking south-east up the sloping plateau of the Aorere Valley. After a photograph by the Geological Survey.

The tilted plateau of the Aorere Valley has been revealed owing to the removal from its surface of the covering strata, a few residual areas of which testify to their former wide extension. Where seen the floor beneath the covering strata proved to be a cleanly eroded surface of fresh rock, and the basal beds in this locality have always been described as conglomerate of marine origin, generally a few feet in thickness and thoroughly sorted, only the hardest of the materials of the undermass, as a rule, surviving. Within a few feet vertically this quartz conglomerate passes into pure limestone, the presence of which indicates prevailing open water, the shore-line being some distance away and any neighbouring land being of small relief and supplying a negligible quantity of mechanical waste.

Bell regards the formation of the denudation plain as having taken place in a strait necessarily initially narrow and afterwards opened by marine erosion to the width of the sloping plateau of the present day, the shores of which strait constituted an “old land” of mature mountains. There are, however, several rather serious objections to this explanation.

First, if the land-forms of the region prior to submergence were mature mountains, and if submergence of such a region of strong relief

– 68 –

led to the formation of a strait, it is difficult to believe that the feeble waves of the waters of a strait, which was necessarily initially narrow and, at least in parts, almost or completely landlocked, can have reduced mature mountains of resistant rocks to a plain of marine erosion, over a width of several miles, in a period so short that the mountainous relief of the “old land” was not destroyed in neighbouring areas by subaerial denudation.

Secondly, submergence of a region of mature mountains must result in the drowning of many valleys. Thus not only would a single narrow strait be formed, but also many branching bays of considerable depth. If we suppose planation of the partially drowned ridges, the peninsulas, and the islands of such a region to take place, it is apparent that the result cannot be a continuous denudation plain. While a more or less perfect plain will have been produced, and this end will have been achieved by the cutting-down of the salient features and the filling-up of the drowned valleys to a common level. The sloping plateau is, however, a plain of denudation throughout. Had it been otherwise the plateau in its present form could not have survived, but the ancient filled valleys would have been re-excavated by modern erosion, revealing again the postulated maturely opened valley forms and the ancient drainage pattern.

Thirdly, a strait the floor of which is a platform of marine erosion must have been much widened by wave-action, and if it has been so widened in a region of mature mountains it must be bordered by wave-cut cliffs of great height. No such cliffs, however, have been pointed out by Bell. It is true that a scarp, previously referred to as the Wakamarama fault-scarp, bounds the Aorere Valley on the northwest side, and it is true also that steep mountain-slopes ascend from the sloping plateau on the south-east side, but recourse has never been had to marine erosion to account for these, and other and more satisfactory explanations are not difficult to find.

Fourthly, the nature of the sediments of the covering strata of the Aorere Valley has been previously referred to. The limestone which follows the basal conglomerate is free from admixture of terrigenous material, and is not the kind of deposit that might be expected to occur in a strait between mountains. It has, moreover, not been shown to pass into a littoral facies towards either side of the supposed strait.

Erosional, Sedimentary, and Deformational History

It has been necessary to state the objections to Bell's theory of the genesis of the physical features of the Aorere Valley somewhat fully, since they are all arguments in favour of the hypothesis which is here offered in its place. It would appear that the strong relief which the deformed undermass presumably had in some earlier period had been almost completely destroyed prior to the deposition of the covering strata. It is reasonable to suppose that this reduction of the ancient mountains was effected largely by subaerial erosion, though planation was completed, over at least the area of the Aorere Valley, by the advancing sea at the commencement of the period of deposition of the covering strait. Next followed a period of deposition over a wide area, and, later, the episode of strong differential movements, which sketched out the broad outlines of the land-forms of the present day, led to the formation of many consequent rivers and inaugurated the cycle of erosion in which the majority of the details of the surface were developed.

– 69 –


The sloping plateau is crossed, as noted above, by a number of young gorges. These are evidently superposed consequent ravines, for they descend the slope of the tilted plateau, and the courses of the streams which cut them were evidently guided by the slope of the surface of the former cover. They are quite indifferent to both dip and strike of the undermass rocks upon which they now flow. These gorges are, in general, narrow-floored and steep-walled, and some distance back from their debouchures the larger of them are incised to a depth of several hundred feet below the sloping plateau. The larger streams head in a range of mountains to the south-east, some peaks of which rise to heights of 2,000 ft and more above the plane of the sloping plateau produced in that direction. These mountains, the Haupiri Range, may be satisfactorily explained as the dissected remains of a higher block separated from the sloping plateau block by a fault or flexure. The streams rising in the Haupiri Mountains have already considerable volume when they begin to cross the sloping plateau in extended, superposed consequent courses, and it is to the action of these vigorous streams that the somewhat advanced dissection of the plateau is to be ascribed. The diagram, Fig. 5, represents the dissection of a sloping block surface such as that of the south-east side of the Aorere Valley by streams from a higher block behind it. A moderate area of the stripped plateau is represented as surviving, and also a few remnants of the cover. On the sloping plateau of the Aorere Valley a few such residuals, small limestone mesas and buttes, remain (Bell, pp. 2425).

The Actual Valley of the Lower Aorere.

In the foregoing sections the name “Aorere Valley” has been used with reference to a tectonic depression (which, as will have been gathered from the description, is strictly neither a Graben nor a fault-angle depression, but partakes of the nature of both), modified as it is at the present day by erosion. Nothing has been said of the Aorere River beyond the fact that it occupies the depression.

It must be noted, however, that the present open valley is largely the work of the river, the amount of waste which it has transported seaward since the episode of differential movements initiating the depression being truly enormous, and consisting of almost the whole of the covering strata (of unknown thickness, but probably several hundred feet at least), as well as a very large contribution from the dissection of the undermass by its tributaries and its own upper course. The actual erosive work performed by the river itself has been considerable, as is evidenced by the occurrence of terrace after terrace up to a height of many hundred feet on the sloping plateau. As these terraces occur on the lower portion only of the sloping plateau, and as the form of the latter has been modified by them only to an inconsiderable extent, they have not been mentioned in the account of the plateau given on earlier pages. When, however, the valley-side is examined closely, the terraces and their thick covering of gravel are readily recognizable. As the gravel is auriferous, it has been largely excavated, and the workings reveal some former channels of the river refilled when from time to time degradation gave place to aggradation.

The terraces occur up to a height of perhaps 600 ft.* above the present level of the river, and there are also residuals of gravel-covered

[Footnote] * Accurate data are not available.

– 70 –

floors at various heights forming flat-topped and terraced hills, which have been noted by many observers, throughout the three- to four-mile-wide valley lowland, indicating that the Aorere has wandered widely on a broad flood-plain more than once during the excavation of its valley in the covering strata of the fault-angle. It would appear that, after the movements of deformation by which the major topographic features were blocked out, the whole region stood some hundreds of feet lower than at present, and that while the land was in this attitude there occurred the great denudation which resulted in the removal of a great part of the covering strata from the upland areas and the carving of the blocks into forms approximating to those of the present day. It is possible that the sea entered the north-eastern end of the fault-angle, but, if such was the case, the re-entrant so formed was no doubt rapidly filled up, and at the same time a considerable amount of aggradation must have taken place throughout the lower parts of the depression. The traces of any such filling have been since removed by erosion. It is a safe assumption that towards the end of the great denudation the Aorere River was not only graded, but had developed a flood-plain with a width of about four miles, of which the highest terraces are remnants. The lower terraces and the present valley of the river, with its discontinuous narrow flood-plain, are to be ascribed to excavation by the Aorere and its tributaries during recent intermittent movements of uplift.

The Gouland Downs Depression

As stated on an earlier page, the Aorere-Gouland depression is continued beyond the divide between the Aorere Valley system and the streams flowing westward to the Tasman Sea. The south-westward continuation, though obviously cognate with the Aorere Valley portion, differs from it so much in certain particulars that it demands separate description. The floor, known as the Gouland Downs, is in most respects homologous with the sloping plateau of the Aorere Valley, while the true boundary of the depression on the north-western side is a scarp which, after an interruption, continues the line of the Wakamarama fault-scarp. A subsidiary tilted block, the Slate Range, however, at the base of and parallel with the Wakamarama Range, separates the latter from the nearly level floor of the depression, which is bounded on the north by the fault-scarp front of the Slate Range block.

Picture icon

Fig 7.—The Gouland Downs, showing the Slate Range on the right Angle of view, from south to west-north-west.

The Floor

The surface of the “downs” plateau (with an area of about twelve square miles) is a plain of erosion similar to that forming the sloping

– 71 –

plateau of the Aorere Valley, with here and there small mesas of covering strata—pure limestone passing downward into a quartz grit with a calcareous cement, with a thin layer of conglomerate at the base. The limestone mesas are riddled with caves, and it is obvious that the removal of this lowest stratum of the cover is being effected mainly by solution. As the limestone areas are forested, they show out conspicuously in contrast with the rest of the plateau, which is bare of vegetation, with the exception of rushes and a few tussocks of coarse grass struggling for existence in a “sour” and slimy soil (see figs. 7 and 8).

There are areas many acres in extent which are quite flat and nearly level; but the surface of the plateau as a whole is by no means uniform. Besides a number of narrow gorges cut recently below the general surface, but collectively not affecting a large area of the plateau, to which reference will be made in a later section dealing with the drainage and dissection of the “downs,” the principal irregularities are such as may be ascribed to deformation of the denudation plain—of course, along with its cover—at the period when the larger differential movements were also taking place. Owing to the presence of a system of small faults, or possibly flexures, generally transverse in direction to the general elongation of the depression, the surface of the “downs” descends towards the middle of the northern boundary in a series of broad irregular steps, each differing in height from its neighbours by a few tens of feet. These may be seen in Plate IV, Fig. 1, which is a view looking south-westward across the “downs” A wider panorama from about the same point of view is shown in Fig. 7. The foregoing appears to be the most satisfactory explanation of the irregularities in the floor of the depression, but it must be remembered that the surface has been long subject to erosion, that an unknown thickness of cover has been removed from it, and that the initial forms of the small fault-line scarps, if such there be, have been much modified. Moreover, the stream which has effected the removal of the debris of the covering strata has wandered rather widely over the area, and as a result there are some more or less definite fluviatile terraces cut on the undermass in the lower central part of the “downs” A layer of river-gravel occurs on, and proves the origin of, the more definite terraces bordering for some distance the stream (the head of the Big River) which now drains the “downs,” and a sprinkling of gravel over a much wider area—perhaps sporadically over the whole floor—may have a similar origin; but it is probable that most of the surface gravel is a residuum of the conglomerate at the base of the former cover. The presence of scattered knolls of the relatively weak limestone of the cover proves that the higher flat areas are not

– 72 –

to any appreciable extent the result of recent lateral planation by streams on the resistant underlying rocks, and indicates quite clearly that they are portions of a stripped floor. The dissection of the floor by streams, though not far advanced, obscures the minor tectonic features of the relief to some extent, for the streams can in no case be consequent on the form of the floor. Theoretically we may expect to find, in addition to insequent, and possibly subsequent, streams, a drainage pattern superposed from the cover, the streams of which, even if wholly consequent on the form of the cover, may be to a great extent indifferent to minor breaks in the floor.

The general form of the floor of the Gouland Downs depression may, if we neglect the minor irregularities referred to above, be described as sloping gently in the form of a half-basin from the east, south, and west against the scarp of the Slate Range, which forms the northern boundary. The lowest part of the basin is at a height of about 2,000 ft above sea-level. On the western and southern sides the plateau slopes gently up to an even sky-line at a height of about 3,000 ft, beyond which lies the valley system of the Heaphv River.

The Eastern Boundary

Picture icon

Fig. 8.—View looking north-east across the Gouland Downs and the “catenary” saddle. Mount Perry is on the right, and the Wakamarama fault-scarp in the distance in the centreAngle of view, about 75°.

Towards the eastern boundary the slope of the surface becomes steeper, as though passing in an anticlinal form, if produced, over Mount Perry and other peaks at the northern end of a range about 4,000 ft in height which bounds the depression on the east. Mount Perry is seen to the left of the centre in Plate IV, Fig 2, and on the right in text Fig. 8. In both figures the rise of the plateau surface with increasing steepness towards the range may be noted. The preservation of the surface on slopes of considerable steepness is explained by the fact that the rocks are indurated shale and quartzite, offering great resistance to erosion. Naturally, the stage of dissection of the slopes rapidly approaches maturity as the steepness increases, until, on the flank of the

– 73 –

range, the denudation plain is completely dissected away, and only fully mature forms developed in the current cycle are seen. So far as the writer is aware, no remants of the denudation plain or of the covering strata are preserved on the higher parts of this range. There can be little doubt, however, as to the general truth of the foregoing explanation of Mount Perry and neighbouring peaks—namely, that they have been carved by erosion, in the cycle still current there, from an upfolded mass of the older rocks, the upper surface of which was, in all probability, prior to the deformation, a denudation plain continuous with that preserved on the Gouland Downs, and carried, like it, a cover of younger strata. Several miles farther south, however, where the same range still forms the boundary of the “downs,” the fold structure of the mountain-flank may be replaced by a fault; but as the range is there composed of granite which is somewhat easily decomposed, and as the slopes are forest-clad, the interpretation of the scarp is not a simple matter.

A “Catenary” Saddle.

The southward-facing scarp of the Slate Range was referred to above as forming the northern boundary of the plateau-floor of the Gouland Downs. As the floor rises towards the east, however, its level approaches that of the crest of the Slate Range, the scarp of the latter diminishing in height and finally dying away. The north-eastern part of the “downs” surface would apparently be continuous, therefore, with that of the summit of the range were it not for the fact that it is separated from it by a deeply eroded gorge—that of the Big River. In the neighbourhood of this gorge dissection of the surface is naturally in a somewhat advanced stage, but there is still an accordance of the levels of the interfluves indicating the initial shape of the warped denudation plain. In this—north-easterly—direction the Gouland Downs surface rises gradually to the saddle which separates the Gouland depression from the Aorere Valley.

– 74 –

The saddle is one of the most striking features in the whole district. Viewed either from the south-west—i.e., from the Gouland Downs (see Fig. 8)—or from the north-east—i.e, from the Aorere Valley—it appears against the sky as a perfect catenary curve four miles in length. It sags from a height of about 4,000 ft. at the south-eastern end (Mount Perry) and about 3,750 ft. at the north-western end (the Wakamarama Range) to a height of about 2,500 ft above sea-level in the centre. The catenary form of the curve is so striking that the feature was pointed out to the writer by a resident of the Aorere Valley as an indication of profound glacial erosion. There can, however, be no doubt that it is of tectonic origin. It is as though, during the episode of uplift, while the Aorere-Gouland depression as a whole lagged behind the blocks which now form the ranges to north-west and south-east, a strip here had failed to break away from either side, but had sagged in the middle so as to assume the true catenary form.

The Slate Range

Mention has already been made of the Slate Range subsidiary block. For three miles—the full length of the block—its scarp forms the northern boundary of the Gouland Downs, while its width in a north-south direction cannot exceed one-third of its length. Unfortunately, a complete description of this interesting little block cannot be given, as the writer saw only the southern side.

At the eastern end, as previously noted, the upper surface of the Slate Range would but for the ravages of erosion be continuous with the higher north-eastern part of the Gouland Downs; but on all other sides the block appears to be bounded by dislocations. The scarp facing the Gouland Downs has been referred to above as the “front” of the range. An assumption has thus been made that the Slate Range is a tilted block, and that opposite to the fault-scarp front facing south there is a back slope to the north. This must be so, because the evenness of the crest-line and the small dissection of the front show that in the vicinity of the crest the upper surface is nearly flat and slopes back so as to lead the drainage northward. A glimpse of the top of the range caught from the slope of Mount Perry confirms the above view. The flat surface of the top of the range can only be a portion of the dislocated denudation plain found throughout the district, and it seems probable that initially the northward slope of the surface and its probable cover formed, with the scarp of the western Wakamarama Range, a fault-angle depression determining a consequent east-west reach of the Big River (see Fig 3), now, no doubt, superposed on and deeply sunk in the undermass.

The southward-facing fault-scarp of the Slate Range, which has an average height of about 700 ft, presents the usual appearance of blunt spurs ending in line (see Fig 7. and Plate V, Fig 1). The spurs and the intervening steep-grade gullies are forested, with the exception of one spur, which stands out as a bare and also sharp-edged facet because its surface is veneered with a thick vein of quartz. The quartz vein evidently filled an ancient fissure which has guided the more modern fault.

The Drainage and Dissection of the Gouland Downs

The drainage of the Gouland Downs is collected in the fault-angle depression between the gently sloping “downs” surface and the scarp

Picture icon

Fig 1—View looking south-westward across the Gouland Downs
Fig 2—Mount Perry and the eastern margin of the Gouland Downs.

Picture icon

Fig 1—The scarp of the Slate Range View looking north-north-west.
Fig. 2—The gorge of the Blue Duck Creek, incised below the plateau surface of the Gouland Downs.

– 75 –

of the Slate Range, the greater part making its way out eastward as the Big River, and the remainder westward as the Saxon River, which joins the Big River before reaching the sea. These east-and-west stream-courses are obviously consequent. The eastward-flowing reach of the Big River is of great interest, as the stream meanders in full maturity upon a wide flood-plain cut but little below the surface of the denudation plain of the “downs,” indicating that the cycle initiated by the great differential movements is still current. Rejuvenation of the valley of the Big River, due to the later movements of regional uplift which have affected this part of New Zealand, has not yet proceeded so far up-stream as to modify the form of the Gouland Downs.

A number of streams which cross the western “downs” in a northward and north-eastward direction to join the streams in the fault-angle are obviously superposed consequents as they follow the general slope of the surface. They cross obliquely the outcrops of the strata of the oldermass which have a uniform north-north-west strike. The streams are roughly graded, though still in narrow, steep-walled gorges, and hence are sunk most deeply beneath the plateau in their middle courses Plate V, Fig. 2, illustrates the type of features thus produced. The sharp contrast between the steep walls of the gorges and the level plateau above is very striking. As the floors of the gorges are occupied to their full width by the streams, the latter rise rapidly, and become impassable after a shower of rain.

Farther to the east the streams flowing towards the Big River cross the “downs” in a north-north-westerly direction—a direction more northerly than that of the general slope of the surface. They have perhaps been guided by irregularities of the initial surface, but it is noticeable that they are parallel with the strike of the strata of the oldermass. Dissection is here more advanced than it is farthest west, and the numerous longitudinal gullies are separated by rounded quartzite ridges suggesting a subsequent origin.

The eastward-flowing consequent reach of the Big River is connected with a westward-flowing consequent reach farther down-stream (see Fig. 3) by a northward-flowing reach, where the stream makes its way in a gorge around the eastern end of the Slate Range. There can be little doubt that this is of (superposed) consequent origin also. The southwestern rim of the “downs” basin is high, and the probable former extension of the surface to the south-westward would be still higher. Thus the present outlet of the Big River may well mark the position of the lowest gap in the rim of the basin-shaped initial surface of the covering strata, some hundreds of feet above the site of the Gouland Downs.

The outlet gorge has been cut to a depth of many hundreds of feet in the extremely resistant rocks of the oldermass. During the process of gorge-cutting the local base-levels on the Gouland Downs have been very slowly lowered, and thus perfect conditions have been afforded for the stripping of the cover from the plateau.