Coastal History of Southern Westland and Northern Fiordland
By C. A. Cotton
[Received by Editor, January 27, 1955.]
The coast of southern Westland may be described as “compound”, in that various processes have contributed to its development and that it has passed through a succession of episodes of coastal history since the inception of an outline not far distant from, and similar in a general way to, that of the shoreline of to-day. Essentially the episodes are five in number: (1) monoclinal tectonic development; (2) morainic progradation; (3) transverse deformation (attributable to compression built up by, or associated with, movement on the Alpine Fault); (4) glacial excavation of fiord-like valleys; (5) progradation of embayments, filling of fiords, and a general smoothing of the outline. These events are independent of, or supplementary to, the changes which must have been brought about by downward and upward movements of ocean level resulting from the Pleistocene world-wide glacierizations.
Many coasts have a long and complicated history, the early stages of which are obscure and even indecipherable. Only a few coasts can be attributed to a recent shift of the shoreline through a considerable distance to a position on the former sea-floor or former land so that they may be described as newly formed and so present no classification problem. Those with an earlier history as coasts naturally fall into Johnson's category of “compound” shorelines (or coasts)— a category that is practically omnivorous because of the changes (large and small) in the positions of shorelines brought about by oscillations of ocean level, especially those which have accompanied the Pleistocene glacierizations and deglacierizations of the earth.
Quite exceptional local conditions make it possible to trace the history of shoreline changes in the compound coast of south-western Westland far back from a present rather simple outline through stages of diversity to a primitive condition that seems also to have been one of shoreline simplicity.
The description “compound shoreline” should, according to its originator, Johnson (1919, p. 191), “be employed … when there is a very marked development of the features characteristic of two or more of the simpler classes of shoreline.” Johnson had in mind especially a combination of features due to submergence with those of prior or subsequent emergence, as a result of the occurrence of “oscillations in the level of land or sea”; but he cited also as compound a possible association of features due to drowning with a contemporaneously developed fault coast. It is clear that he regarded the description “compound” as applicable to coasts that exhibit contrasting features, due in some cases to dissimilar processes, whether these have been developed simultaneously or successively. In rather exceptional cases such features may preserve evidence of a series of consecutive episodes in the geological history of the
coast. In the case of the Westland compound coast features are incorporated that record the following succession of phases: (1) a monoclinal coast; (2) a stage of morainic progradation; (3) development of transverse deformation (or buckling) with formation of large embayments separated by tectonic salients; (4) fiord development; (5) and finally progradation and rectification of the outline by marine beach-building in association with alluvial accumulation.
These stages are independent of the contemporaneous oscillation of ocean level of considerable amplitude that is known to have taken place and can scarcely fail also to have left some traces; and, further, earth movements of buckling—attributable to compression—cannot have been confined to stage 3, for transcurrent-fault movement parallel to the coast—a likely cause of buckling in the coastal strip, or foreland—has been in progress for a long time and still continues (Wellman and Willett, 1942A; Wellman, 1952; Munden, 1952).
Fig. 1.—Monoclinal-flexure coast at the mouth of the Paringa River. View looking east. (From a photograph.)
The ancient monoclinal-flexure coast (stage 1) survives in places, or rather has been restored after an indefinite number of sea-level oscillations. It is seen, for example, at the mouth of the Paringa River (Fig. 1).
Order of Events
The Coastal Monocline
Evidence of a strong monoclinal flexure is now afforded by a narrow marginal belt of mid-Tertiary strata with steep seaward inclination (Fig. 2).
The precise date of the origin of the boundary between land and sea by development of the monoclinal flexure is not known, but its formation certainly must have been delayed until after the period of deposition of the youngest of the strata involved in the flexure. The actual shoreline on the belt of Tertiary rocks is indeed structural, not tectonic, for, as Wellman (1951, p. 19) points out, it follows the strike of the rock strata so closely that it must be inferred that marine erosion has removed an unknown thickness of softer younger-Tertiary beds (overlying those now exposed) in the process of development of
the existing bold and rocky coast along the outcrops of the well indurated Tertiary strata that survive. This, however, is a development of detail due to erosion partly in the current marine cycle, as well, almost certainly, as during former high stands of sea-level between withdrawals in glacial ages, and it does not prejudice the classification of the coast as primarily monoclinal.
There can be little doubt that the monocline originated at some time during the Kaikoura deformation of late-Tertiary and post-Tertiary date, when not only the major relief features of New Zealand (associated with development of some complexity of geological structure) had their origin but also the broad outlines and even some locally surviving minor elements of the coastline were determined.
Though the evidence of monoclinal flexure on a gigantic scale is to be found especially in the steep seaward dips observed in the marginal remnants of mid-Tertiary strata (Fig. 2), which must of course have formerly been far more extensive, this structure must flex also the underlying complex of older formations which is the undermass of the foreland belt. Clearly the structure of the foreland is not simply homoclinal but is monoclinal—i.e., affected by a monoclinal flexure. It seems probable that this flexure is continued south-westward along the curving tectonic coast of Fiordland.
Vast morainic accumulations—termed the Piedmont (or Paringan and Kinnaird) moraines by Wellman (1951, p. 30)—reach the coast for a long stretch to the north-east of the area mapped in Figs. 2 and 3. They are attributed to
Fig. 2.—Essential features of the geology of south-west Westland and northernmost Fiordland, showing the trace of the Alpine transcurrent fault, evidence for the coastal monoclinal flexure, and some deposits affording clues to Pleistocene and postglacial history. (After Wellman and Willett, 1942A, and Wellman, 1951.)
Fig. 3.—Sketched restoration of the shoreline of early postglacial time. (After Wellman and Willett, 1942B.)
Fig. 4.—Schematic representation of features related to five stages of coastal development. At right developmental features referable to stages 1-4 are shown (see text); at left, the condition of the present-day coast, stage 5.
intense glaciations of the region long prior to the milder one (Alpine, of Wellman) in which fresh-cut fiord valleys and in general the surviving glaciated topographic features of the Southern Alps were shaped.
Within the area shown in Fig. 2, the coastal salient of morainic debris at Cascade Point is attributed by Wellman and Willett (1942B) to the same stage as the Piedmont moraines of other parts of the coast. This correlation appears to be correct as regards the bulk of the material in the Cascade salient, including the thick mass of coarse material interpreted by Turner (1930) and by Benson, Bartrum, and King (1934) as preglacial conglomerate, though there is apparently a veneer of younger moraines on its flattish top (Cascade Plateau).
Buckling (with Production of a Coast of Transverse Deformation)
It is not quite certain that the buckling indicated by presence of horst-like or possibly arched salients seaward of the Alpine Fault, which are separated by (former) embayments (Figs. 2, 3, 4), took place subsequently to the stage, described above, of morainic progradation—i.e., the Piedmont glaciations. The absence of morainic infilling from the Haast embayment, or, quite probably, its submergence there below the present sea-level, suggests, however, the order of events here tentatively assumed. The relation of some beach deposits, of suggested “early Pleistocene” but really of undetermined age, at altitudes of several hundred feet (Wellman and Willett, 1942B, p. 217) to the Piedmont moraines, the glacial debris of which overlies them and is not itself apparently benched by marine erosion, might be taken to indicate that differential movement (upheaving beach deposits) took place prior to the Piedmont glaciation; but this may be explained by assuming that there was more than one episode of differential upheaval. Such a conclusion is inevitable, indeed, on a priori grounds, for the great transcurrent fault parallel to the coast (Fig. 2) is regarded by geologists as a dislocation that has long been active, and during its activity the foreland strip must be continuously subject to compression and therefore unstable vertically, or liable to local upheaval and depression. In Fig. 4 a schematic picture is presented of a postulated buckling of the foreland strip—the strip between the Alpine transcurrent fault and the coastal monocline. Such buckling is postulated especially to explain “stage 3” shorelines, which are due to an apparent downwarping or downfaulting of the Haast lowland and of another similar depressed area, about 20 miles wide, between the end of the high coast near the Paringa River (Paringa horst) and the Cook River, to the north-east.
The transcurrent-fault fracture, like others of its kind, undoubtedly extends to a very great depth, and the buckling of the strip seaward of it may be regarded as affecting the crust to a great depth also. It cannot be a superficial effect, but must be rather of the nature of plis de fond. Deep-seated large-radius folding of such a kind—a miniature of that postulated (on an immense scale) to account for the superficial structures of Eurasia (by Argand, 1924) and of South America (by Ruellan, 1953)—may have as its surface expression warping or block faulting, or, to judge from examples in other parts of New Zealand, some combination of these, the result resembling deformation associated with Germanotype, or Saxonian, structure. That local development of small-scale plis de fond still continues, accompanying the known continuance of activity of transcurrent movement as demonstrated by Wellman (1952) and Munden (1952), may be deduced from the presence along sixteen miles of rocky coast mainly south-west
of the Paringa River of a narrow, quite freshly developed, erosional rock bench which is tilted endwise (downward to the south-west end, where it reaches sea-level) giving an indication that differential upheaval of the Paringa horst, as it may be termed, is still in progress (Cotton, 1947, p. 372). (Only a small part of this minor feature, that reported by Wellman, 1951, p. 9, at altitude 50ft, is indicated in Fig. 4.)
Upheaval and folding of marine Pleistocene strata associated with Piedmont morainic debris, which are exposed in the Paringa Valley (Wellman, 1951, p. 23), may be attributed to development of plis de fond in the foreland. In view of the evidence of differential movement that may be so controlled it seems unwise to extrapolate from the graph drawn by Wellman (1951, Fig. 2) from Paringa Valley data of fluctuation of Pleistocene sea-level “above datum”, though a regional generalization from this for the West Coast has been suggested (Wellman, 1951, pp. 27-8).
A stage of glacial-trough and fiord development is assigned by Wellman and Willett (1942B, pp. 215-7) to the latest phase of glaciation, the Alpine of Wellman (1951), to which also the excavation of the freshly cut inner trough of Milford Sound is attributed by Bruun, Brodie, and Fleming (1955). Wellman and Willett have sketched (as copied in Fig. 3) shorelines of early postglacial time (perhaps of only 6000 or 7000 years ago) on the assumption that the fiords of Fiordland and numerous filled fiord troughs in northern Fiordland and more particularly south-west Westland were eroded to their present shapes in the latest period of low-level glacial erosion. It is well known, of course, that troughs of great depth below present sea-level can be thus explained without necessarily assuming changes in level of land or sea other than that due to the return of the ocean more or less to its former level after the lowering of sealevel that accompanied the last major glacierization of the earth.
Progradation of Tectonic Embayments and Infilling of Fiords
The broad belt of lowland (surrounding former islands) in the graben-like Haast embayment and other strand plains to the north-east of the Paringa horst (Fig. 5) are of obviously recent development in relation to sea-levels close to that of the present day. They consist of gravelly alluvial plains and swamps bordered seaward by prograded areas consisting of dune tracts and forelands built of successively formed beaches of sand and gravel. Wellman reports, for example, a 14-mile-long gravel beach at the north-east end of the Haast embayment. “The north-eastern part [of the beach] is separated from lakes farther inland by a wide belt of sandhills. Air photographs show six subparallel sand ridges separated by swampy depressions which gradually diverge [south-westward]. These ridges were formed during the intermittent advance of the coast, which swung out as though hinged at [the north-east end], each advance bringing the coast more nearly in line with the sweep of the rock-controlled coast [of the Paringa horst] farther north [-east]. The oldest and most inland sandhill marks the position of the sandspit which first converted the one-time seafilled [Haast] depression … into a fresh-water lake. Most of this lake has since been filled by the growth of alluvial fans” (Wellman, 1951, p. 9).
Argand, E., 1924. La tectonique de l'Asie, C. R. XIII Cong. géol. internat. (Brussels), vol. 1, pp. 171-372.
Benson, W. N., Bartrum, J. A., and King, L. C., 1934. The geology of the region about Preservation and Chalky Inlets, Trans. N.Z. Inst, 64, pp. 51-85.
Cotton, C. A., 1947. The Alpine Fault of the South Island of New Zealand from the air, Trans. Roy. Soc. N.Z., 76, pp. 369-372.
Johnson, D. W., 1919. Shore Processes and Shoreline Development, New York: Wiley.
Munden, F. W., 1952. Notes on the Alpine Fault, Haupiri Valley, North Westland, N.Z. Jou [ unclear: ] . Sci. Tech., 33B, pp. 404-408.
Ruellan, F., 1953. Le rôle des plis de fond dans la structure et le relief du boucher sudaméricain, C R. XIX Cong. géol. internat (Algiers), vol. 3, pp. 241-261.
Turner, F. J., 1930. The physiographic features of the lower Cascade Valley and the Cascade Plateau, S. Westland, Trans. N.Z. Inst., 61, pp. 524-535.
Wellman, H. W., 1951. The geology of Bruce Bay-Haast River, S. Westland, N.Z. Geol. Surv. Bull., 48.
—— 1952. The Alpine Fault in detail: river terrace displacement at Maruia River, N.Z. Jour. Sci. Tech., 33B, pp. 409-414.
—— and Willett, R. W., 1942A. The geology of the West Coast from Abut Head to Milford Sound, Part 1. Trans. Roy. Soc. N.Z., 71, pp. 282-306.
—— 1942B. The geology of the West Coast from Abut Head to Milford Sound, Part 2 (glaciation), Trans. Roy. Soc. N.Z., 72, pp. 199-219.
Prof. C. A. Cotton2 Manuka Avenue