The Origin of the Opouawe River and Nearby Antecedent Gorges in S.E. Wairarapa

Transactions and Proceedings of the Royal Society of New Zealand, Volume 87, 1959, Page 91

The Origin of the Opouawe River and Nearby Antecedent Gorges in S.E. Wairarapa By J. B. Waterhouse, Geology Department, Victoria University, Wellington [Received by the Editor, October 30, 1958.] Abstract Several consequent streams that flow from the Aorangi Mountains towards the east coast have been defeated by the uplift of Ewe Ridge across their courses. The defeated streams have joined along a synclinal depression parallel to the ridge to form the Opouawe River. Downstream from Ewe Ridge the streams before defeat had carved antecedent gorges through a coastal range, and these gorges are now occupied by underfit streams that flow in the former lower courses of the intercepted streams. Introduction It is common in New Zealand to find streams and rivers that have carved valleys through a range instead of taking an “easy” way round the obstacle. Particularly close attention has been paid by King (1934) and Cotton (1938) to the origin of gorges through a range in front of the Haldon Hills of north-eastern Marlborough, and the origins of other New Zealand gorges have been discussed and summarised by Cotton (1948, p. 146). Yet another kind that does not belong to any category hitherto described in New Zealand is apparently represented in south-eastern Wairarapa. Generalised Account of the Structure and Landscape In the south-east corner of the Wairarapa district there are three ranges running north-east and south-west, these being from west to east the Aorangi (or Haurangi) Mountains, 3,000ft high, Ewe Ridge, up to 1,500ft high, and an unnamed coastal range, up to 1,200ft high (Fig. 1). Structurally the two westward ranges are similar in that each is developed from an asymmetrical anticline with a faulted south-east limb. Perhaps the coastal range is similar in structure, but little is known of its south-east limb, for this lies unexposed beneath the sea. The rock of the Aorangi Mountains comprises chiefly? Jurassic and Lower Cretaceous greywackes; Ewe Ridge comprises Lower Cretaceous greywackes of the Urewera Series of Wellman (1956), and in the coastal range the rock consists of Upper Cretaceous (Mata) and Lower Tertiary limestones, greensands, and redeposited shelf sediments, as described by Waterhouse and Bradley (1957). Between the Aorangi Mountains and Ewe Ridge is a lowland, varying from two to three miles in width, and sloping gently from the foot of the mountains to a synclinal depression near the foot of Ewe Ridge. In the lowland are greywackes, some similar in age to and some a little younger than those of Ewe Ridge, together with Miocene cover. The lowland between Ewe Ridge and the coastal range is up to a mile wide, with Lower Tertiary and Upper Cretaceous strata folded in a syncline. The south-east face of the Aorangi Mountains presents an imposingly steep scarp with a relief of 2,000ft, developed from the two major faults that bound the south-east flank of the range. From the fault scarps valleys extend back for a mile

Fig. 1.—The Opouawe and Awhea river systems, SE Wairarapa. Tributary “a” of the Opouawe River has captured the headwaters of Oroi Stream. and a half into the range. In contrast the south-east fault scarp of Ewe Ridge is still fresh and little eroded, with streams extending into the range for no more than 16 chains. The extended north-west flanks of both ranges are dissected by consequent streams, between which the divides show a rough accordance of summit level. In the coastal range the north-west flank is moulded from a cap-rock of resistant limestone (the Mungaroa Limestone of Waterhouse and Bradley, 1957, p. 524), and the south-east face has been pared back by the sea to high cliffs with narrow remnants of marine platforms. Generalised Account of the Drainage Opouawe River and Its Tributaries From the southern Aorangi Mountains five large streams flow in a south-easterly direction towards Ewe Ridge. They are roughly parallel to one another and are of consequent origin, for their courses were initially guided by the slope of the lowland

east of the mountains. Of these streams the northern four—namely, the Cape, Rough, Castle, and Poley, flow into the Opouawe River, which is also of consequent origin, flowing in or close to the synclinal depression between the lowland and Ewe Ridge. Synclinal control is most pronounced in the upper reaches of the river, where the lowland and the extended north-west limb of Ewe Ridge are strongly inclined towards each other. Downriver the Opouawe flows to the east of the synclinal axis, in the foot-hills of Ewe Ridge. Presumably the river took this course in the past when Ewe Ridge was narrower and the topography lower, and then, at a later stage when Ewe Ridge was further uplifted and its base extended westwards beyond the river, the river entrenched itself within the foot of the range. A somewhat similar phenomenon is described by Barrington Brown and Debenham (1929) in New South Wales, where the Nepean River has become incised into the lower slopes of the Blue Mountains. Both Ewe Ridge and the coastal range extend little south of Poley Stream, and the river swings around the foot of Ewe Ridge to enter the sea at the south end of the coastal range (Fig. 2). At the south end of Ewe Ridge the river is superimposed on to a tight fold of Teurian limestone (in the Manurewa Formation of Waterhouse and Bradley, 1957, p. 522), lying on the downthrown side of the Ewe Ridge fault. To the south, at White Rock, the same fold acts as a barrier to Whawanui Stream, turning the stream aside from its consequent course and forcing it to flow in a south-westerly direction to the sea. Fig. 2.—Looking from the south end of Ewe Ridge towards the mouth of the Opouawe River. The south end of the coastal range (c) lies to the left. In the middle of the sketch is a ridge of folded Teurian limestone (a-a), across which the river has become superposed. Lower Cretaceous greywackes lie immediately to the right. It will be noticed from Fig. 1 that the Castle and Rough streams join before meeting the Opouawe River. This change in course is possibly of consequent origin, as the junction of the two streams is close to the axis of the synclinal depression. From Ewe Ridge, the Opouawe receives a number of small subequal tributaries, some of them too small to be shown on the map. Although these are at a comparatively early stage of consequent development sufficient time has elapsed to allow some subsequent modifications, for tributary branches have developed headwards along less resistant bands of strata.

Streams that cross the coastal range Between the Opouawe River and the east coast are three small but remarkable streams, the Oroi, Pukemuri, and Awheaiti, that arise on or near the south-east scarp of Ewe Ridge. The way to the sea is barred by the coastal range, reaching hundreds of feet above the lowland, but the streams cross the range each in a separate gorge. These streams are tiny; even during floods their volume is insignificant. Only the Awheaiti is now at grade, probably because it crosses a part of the range that has been considerably narrowed by sea-cliffing; there is a waterfall over 100 feet high in the valley of the Oroi, and another about 40 feet high in the valley of Pukemuri Stream. It is significant that whereas degradation occurred during the last glacial episode in the valleys of the nearby Opouawe and Awhea rivers, and was followed by aggradation in the following warmer period when the sea-level rose, these small streams were not sufficiently incised during the cold epoch to have been affected by the ensuing aggradation. Origin of the Gorges Through the Coastal Range Rejection of subsequent or insequent origin for the gorges Theoretically it appears possible that the three small streams once flowed from Ewe Ridge along the lowland and into the Awhea or Opouawe River without crossing the coastal range. Later, it might be proposed, these streams could have been diverted by subsequent or insequent streams that had worked headwards through the coastal range from the shore. Were this theory correct, there should probably be an elbow of capture where each stream was diverted from its former course along the lowland, for it is unlikely that the stream would have been captured exactly at the place where it formerly turned aside from the coastal range. From Fig. 1 it is clear that the Awheaiti and Pukemuri Streams flow directly across the lowland to the sea without elbows of capture. In the course of the Oroi there appears to be a sharp turn, but this is where a large tributary joins the main stream—it is not an elbow of capture. The Oroi Stream now arises in the lowland (Fig. 3), having lost its headwaters from Ewe Ridge to a small subsequent tributary from the Opouawe River (marked in Fig. 1). A wide air-gap is still preserved in the low divide between the pirate stream and the truncated remnant of Oroi Stream. The resistant limestone and other highly compacted rocks of which the coastal range is comprised make it difficult for insequent streams to progress headwards into the range from the coast, so it seems unlikely that such streams could have attained a gradient low enough to have captured streams in the lowland. Moreover, excellent exposures of well-bedded rock in the gorges show that there are no structural weaknesses which might have favoured the development of subsequent streams. Suppositions that insequent or subsequent streams could have worked headwards through the coastal range and lowland to Ewe Ridge without capturing streams must also be rejected for the same reasons. Superposition on to the coastal range In discussing the origin of the gorges carved through a fore-range by streams that flow from the Haldon Hills, Cotton (1938) suggested that the streams may have been superposed on to the fore-range from fans of detritus built from the hinter range. Applied to the coastal range problem, such a theory involves the building of large fans from the fault scarp of Ewe Ridge across the lowland and over the coastal range. Consequent streams would have flowed over the fans from Ewe Ridge and become superposed on to the range as dissection ensued. To have provided sufficient fan-debris to bridge the lowland (which is a mile wide) and bury the coastal range, a great thickness of strata must have once lain

Fig. 3.—Headwaters of the Oroi Stream in the lowland between Ewe Ridge and the coastal range, viewed from the coastal range. In the background is the SE scarp of Ewe Ridge, dissected by headwaters of Pukemuri Stream. Part of the gorge through which the Oroi Stream traverses the coastal range is seen at the lower right corner of the sketch. over the present Ewe Ridge. Yet, during the early phases of the uplift of Ewe Ridge in the present cycle of erosion, the present crest and back-slopes of the range were probably covered by no more than a few hundred feet of greywacke, possibly capped by 100 to 300 feet of Miocene cover. The Upper Cretaceous and lower Tertiary beds, and much Cretaceous greywacke, were removed west of the Ewe Ridge fault from the greywackes of Ewe Ridge and the Opouawe lowland before a Miocene transgression in which the Hurupi and Bell Creek beds were deposited east and west of the Aorangi Mountains, directly over truncated Lower Cretaceous greywackes. On the divides between small tributaries east of Awheaiti Stream there are, in fact, remnants of fans built of Cretaceous greywacke rubble from Ewe Ridge As the fans extend only a few hundred feet from the fault scarp, they are much too small to have bridged the lowland. For these reasons the theory of superposition from fans does not appear to afford a likely explanation for the origin of the gorges through the coastal range. Theory of antecedent origin Less definite objections can be raised against a possible antecedent origin similar to the suggestions made by King (1934) for the gorges through the fore-range of the Haldon Hills. Applying these suggestions to the coastal range problem, Ewe Ridge would have been uplifted first, with the development of consequent streams that flowed without obstruction from the fault scarp to the coast. Later, according to this theory, the coastal ridge was uplifted, and in spite of the indurated strata of the range and the small size of the streams, the streams were able to hold course across the uprising range. As Cotton (1938, p. 193) pointed out in discussing King's theory for the Marl-borough gorges, the rate of uplift must not be too fast, lest the streams be defeated. Although there is no direct evidence to show that the coastal range has not in fact been uplifted at a rate which favoured the formation and maintenance of the gorges, this theory is not favoured in its application to the gorges of the coastal range for the following three reasons.

In the first place, it would seem that a long time must have been required for these small streams to carve gorges across the uprising range through resistant limestone and other strata. Thus time should have been sufficient for subsequent streams to have worked headwards from the Opouawe and Awhea rivers for a considerable distance into the soft rock of the lowland. But there are only a few short subsequent streams, seemingly at an early stage of development. Secondly, the suggested antecedence does not explain how such small streams have succeeded in carving the gorges. These gorges are almost as large as that carved by the much larger Awhea River to the north-east. In each gorge only the lower part, three or four hundred feet high, is narrow and gulch-like, undoubtedly carved by the stream that occupies the valley. Finally, an antecedent origin as outlined above assumes that Ewe Ridge was uplifted before the coastal range. However, it appears likely that Ewe Ridge may be the younger of the two landforms in the present cycle of erosion. Whereas the base of the fault scarp of Ewe Ridge coincides with the line of the fault, the southeast face of the coastal range, whatever its original nature, has retreated for at least a quarter of a mile from the anticlinal axis. Of course this considerable retreat has been speeded by the proximity of the sea, probably a much more active agent of erosion than the forces that are at work on the Ewe Ridge fault scarp, but allowing for this it seems reasonable to propose that the uplift of the coastal range commenced before the uplift of Ewe Ridge. Truncated antecedent origin for the gorges through the coastal range It appears that the problem can be best explained by proposing that Ewe Ridge, the present source of the small streams that cross the coastal range, has been uplifted after the Aorangi Mountains and after the coastal range (Fig. 4). In this favoured hypothesis it is suggested that the Aorangi Mountains were the first to be uplifted, and that this range later became bordered to the east by a low plain with a thin upper Tertiary cover, now largely removed. Guided by the initial slope of the land, streams flowed in a south-easterly direction from the Aorangi Mountains to the coast, crossing the future sites of Ewe Ridge and the coastal range, which were then not apparent as uplands. These consequent streams were the ancestral Whawanui, Poley, Castle, Rough, and Cape streams. Later the coastal range began to be uplifted across the lower courses of the streams. Being large, these streams were not diverted, but were able to hold course and carve antecedent gorges through the uprising range. After the uplift of the coastal range had commenced, Ewe Ridge also began to rise, rapid uplift and the resistant nature of its rock enabling the new range to defeat the consequent streams from the Aorangi Mountains. Thus the streams that crossed the coastal range had their headwaters transferred from the Aorangi Mountains to Ewe Ridge. After truncation much of the volume in the lower courses was lost, which explains why the coastal range streams are now underfit, for they have inherited gorges that were carved for them by much larger streams. Allowing for some clockwise transcurrent movement along the Ewe Ridge fault, each of the streams that cross the coastal range appears to be still roughly aligned with its severed headwater course, Awheaiti Stream matching Rough Stream, Pukemuri Stream matching Castle Stream, and Oroi Stream matching Poley Stream. The early stage of this change in the drainage pattern, in which a low ridge with air-gaps separates streams from their former lower courses, has passed, and air-gaps are no longer preserved in Ewe Ridge. The Origin of the Opouawe River When Ewe Ridge began to rise a depression was formed between the ridge and the lowland east of the Haurangi Mountains. The streams defeated by the rise of Ewe Ridge flowed into this depression, and united to form the Opouawe River, a

Fig. 4.—The favoured hypothesis of origin of the Opouawe River and the gorges through the coastal range. (a) The Aorangi Mountains are uplifted and consequent streams flow southeastward from the range over a low coastal plain. (b) The coastal range begins to emerge. The consequent streams cut antecedent gorges through the uprising range. (c) Ewe Ridge is uplifted between the Aorangi Mountains and the coastal range. The streams from the Aorangi Mountains are diverted to form the Opouawe River west of Ewe Ridge. Small underfit streams flow through the coastal range in the lower courses of the truncated Aorangi streams. synclinal consequent river which is younger than its main tributaries. The Opouawe River was formed so recently that there has not been sufficient time for subsequent streams to develop from the river into the lowland between Ewe Ridge and the coastal range. This helps to explain why the small streams that cross the coastal range remain almost intact from piracy by subsequent streams. Since the formation of the Opouawe River, the increase in volume of the Awhea River (as explained below), and the reduction in volume of the three streams that cross the range, the pattern of drainage has become unstable, and subsequent streams are now likely to effect further changes in the drainage pattern. Because the Opouawe tributaries such as Cape Stream no longer have to cross the uprising coastal range, these now easily maintain grade, thus partly explaining the considerable rejuvenation in the upper courses of these streams. McLean (1953) has described how Cape Stream is extending its headwaters at the expense of the northward flowing Haitai Stream. The Origin of the Awhea River The Awhea River lies to the north-east of the Opouawe River and is of interest in that like many New Zealand rivers it has a right-angled bend in its course. The upper part of the river, named Whakapuni Stream, flows south-west to Tuturumuri to join Stony Creek. The two flow towards each other in a synclinal depression that is clearly the north-east continuation of the Opouawe depression. After meeting they turn abruptly into a south-easterly direction, flowing along a tectonic sag

with infaulted Manurewa beds between Ewe Ridge to the south and Waipawa Hill to the north. The suggested origin of the river is as follows. Assuming that Ewe Ridge and Waipawa Hill were uplifted together, minor streams that had flowed across the uplands would have been defeated, and would have collected in the synclinal depression in the same way as the Cape and other streams were defeated to form the Opouawe River. Instead of flowing southward along the depression into the Opouawe these northerly streams pierced the upland through the tectonic sag at Tuturumuri, conceivably but not necessarily emptying into one of the original consequent streams which may have flowed at this place. For a mile the river, aided by the tectonic sag and reinforced by the defeated streams, flows in this direction. Then the course becomes complicated. Briefly, the river turns southwards to flow along a synclinal depression east of Ewe Ridge for nearly three miles, and then turns into a more easterly direction to cross the coastal range, apparently through the gorge shaped by the lower course of the ancestral Cape Stream. The river is easily able to hold its course through the coastal range, and, unlike the feeble streams to the south, is not likely to be threatened by capture. Other rivers in the Wairarapa district such as the Wainuioru have a right-angled bend in their course, and it is possible that here also defeated streams flow along tectonic depressions into rivers which, favoured by large volume or tectonic sags, are able to maintain course through antecedent gorges. In the Tararua and Rimutaka ranges of the North Island the pattern of drainage shown by the rivers is similar to that of the Wairarapa, and the rivers are probably consequent or anteconsequent as suggested by Adkin (1947). Following Cotton (1957) in his description of the drainage pattern of Wellington Peninsula, it is probable that the synclinal or fault-angle depressions which guided some of the rivers were asymmetrical and faulted (see also Adkin, 1947, Fig. 5), and perhaps the highs arose at different times, and grew at different rates, as in the Wairarapa. Acknowledgments This paper is based on field work by the writer for an M.Sc. thesis in geology at Victoria University. The writer is grateful to Professor C. A. Cotton and Dr. M. T. Te Punga for their generous advice and helpful criticism. Figs. 1 and 4 were drafted with the help of Miss J. Symons at the New Zealand Geological Survey. References Adkin, G. L., 1947. The Tararua Range as a unit of the Geological Structure of New Zealand. Trans. roy. Soc. N.Z., 77:260-72. Barrington Brown, C., and Debenham, F., 1929. Structure and Surface. London. Cotton, C.A., 1938. The Haldon Hills Problem. J. Geomorphy, 1:187-98. —, 1948. Landscape as Developed by the Processes of Normal Erosion. Wellington: Whitcombe and Tombs. —, 1957. Tectonic Features in a Coastal Setting at Wellington. Trans. roy. Soc. N.Z., 84:761-90. King, L.C., 1934. The Geology of the Lower Awatere District, Marlborough, New Zealand. N.Z. Geol. Surv. Mem., 2. McLean, D.B., 1953. The Geology of the Haurangi-Stony Creek Area, south-east Wairarapa, Univ. of N.Z., unpublished M.Sc. thesis, lodged in Victoria University Library, Wellington. Waterhouse, J.B., 1954. Geology of the White Rock-Tora Area, south-east Wairarapa, Univ. of N.Z., unpublished M.Sc. thesis, lodged in Victoria University Library, Wellington. — and Bradley, J., 1957. Redeposition and Slumping in the Cretaceo-Tertiary Strata of S.E. Wellington. Trans. roy. Soc. N.Z., 84: 519-48. Wellman, H. W., 1956. The Cretaceous of New Zealand. 20th Intern. geol. Congr. Mexico, Res. trabajos presentados: 351-2. Dr. J. B. Waterhouse, N.Z. Geological Survey, Lower Hutt