Sedimentology and tectonic setting of the Late Permian-early Triassic Stephens Subgroup, Southland, New Zealand: an island arc-derived mass flow apron

Sedimentary Geology, 68 (1990) 55-74 55 Elsevier Science Publishers B.V., Amsterdam

Sedimentology and tectonic setting of the Late Permian-early Triassic Stephens Subgroup, Southland, New Zealand"

an island arc-derived mass flow apron

J .C. A i t c h i s o n * a n d C .A. L a n d i s

Department of Geology, Unioersity of Otago, 1'. O. Box 56, Dunedin (New Zealand)

Received April 19, 1989; revised and accepted April 20, 1990

ABSTRACT

Aitchison, J.C. and Landis, C.A., 1990. Sedimentology and tectonic setting of the Late Permian-early Triassic Stephens Subgroup, Southland, New Zealand: an island arc-derived mass flow apron. Sediment. Geol., 68: 55-74.

The Late Permian-early Triassic Stephens Subgroup in southern South Island, New Zealand was derived from an active volcanic arc and deposited in an elongate arc-flanking deep-sea basin. The Stephens Subgroup is 2500 m thick and comprises five formations. Units are lateratly extensive and are traced at least 40 km within the region studied. Strata are assigned to five lithofacies which resemble those of modem and ancient deep-sea fans, particularly mid-fan facies. However, there is no evidence to suggest existence of classical radial fan geometry and the possibility of deposition along a submarine ramp is also considered. Sandstones belong predominantly to a quartz-deficient epiclastic volcanic sand petrofacies, but minor vitroclastic tuffaceous sandstones are also recognised. Modal analyses, composition of detrital clinopyroxene and whole rock geochemistry indicate a moderately evolved calc-alkaline oceanic arc source.

Analysis of sedimentological data in conjunction with stratigraphy and structure suggest accumulation of Stephens Subgroup in a Late Permian-Middle Triassic forearc or backarc basin.

Introduction

Volcanogenic sedimentary rocks of the Stephens Subgroup (Altchison et al., 1988) comprise the

upper par t of the Permian-Tr iass ic Maitai Group

(Waterhouse, 1964; Landis, 1980). They are exten-

sively exposed between Mossbu rn and the

Countess Range in western Southland, South Is-

land, New Zealand (Fig. 1). Lithostrat igraphically correlative strata, including the Stephens type lo-

cahty, are well known on the northwest side o f the

Alpine Fault in Nelson, while faul t -bounded slices occur along the Alpine Faul t at Mataki taki and

Gorge Plateau (Fig. 1). Stephens rocks consist predominant ly of sedimentary gravity flow, deep-

* Currently at: Department of Geology and Geophysics, Uni- versity of Sydney, N.S.W. 2006, Australia.

mar ine volcanogenic , epiclastics, with lesser

amounts of tuff, hemipelagic muds tone and lime- stone.

Previous descriptions of western Southland

Stephens rocks have been presented by Aitchison

et al. (1988), bui lding on work by Grindley (1958),

Waterhouse (1964, 1979), Landis (1974, 1980) and

Hyden et al. (1982). In Nelson, the strata are

described by Waterhouse (1964), Johns ton (1981),

Johns ton and Stevens (1978) and Campbel l et al.

(1984). These studies comprise mainly hthologic, stratigraphic and petrographic accounts. The age of the Stephens Subgroup remains controversial in

spite of control provided by physical s trat igraphy and widely scattered bu t extensively documented

faunas. For the purposes of this paper, it is suffi-

cient to state that the upper Stephens (Snowdon

Format ion) rocks are of unquest ionable Triassic

0037-0738/90/$03.50 © 1990 - Elsevier Science Publishers B.V.

56 J.C. AITCHISON AND C.A. LANDIS

m Dun Mountain- Maitai terrane

Brook Street terrane

Caples terrane

Murihiku terrane

Stephens Is~

N

Plateau

Mossburn

Range ~ B a r e Pk { 4s°&s

Dunedin

ii i:i i :: o ,ookm

170OE

Fig. ~. Location map of Countess Range, Snowdon Peak, Bare Peak and other areas of Stephens Subgroup outcrop, South Island, New Zealand. Associated geological terranes (Brook Street, Murihiku, Dun Mountaln-Maitai, Caples) are also shown; note that belts

of Mufihiku terrane adjoining the Alpine Fault are exaggerated in width for clarity.

age while lower Stephens strata fall within the range mid Permian-mid Triassic (Aitchison et al.,

1988). Good exposure, relatively simple structure and

minimal metamorphic reconstitution, particularly in the Countess Range, permit detailed evaluation of the sedimentology and sedimentary petrology

of the Stephens Subgroup. These, along with interpretation of depositional and tectonic setting, form the basis of this study. The paper is based largely on the M.Sc. thesis of Aitchison (1984). Grid references refer to the national metric grid (NZMS 260 sheets D42 and D43). Samples are stored in the Geology Department , University of Otago.

SEDIMENTOLOGY AND TECTONIC SETTING OF THE LATE PERMIAN-EARLY TRIASSIC STEPHENS SUBGROUP, NEW ZEALAND 57

Regional geology

The Maitai Group forms a distinctive and col- ourful suite of Permian-Triassic sedimentary rocks which is restricted to the Key Summit Regional Syncline in western Southland and northwestern Otago and to the correlative Nelson Regional Syn- dine, 480 km to the north across the Alpine Fault in Nelson Province (Fig. 1). Constituent rocks are mainly epiclastic (and lesser pyroclastic) volcano- genie in origin and are sparsely fossiliferous. How- ever, limestones and detritai quartzo-feldspathic sediments are also present (Landis and Blake,

1987) and these tend to be more richly fossiliferous. The group exceeds 6000 m in thickness, its base resting in depositional contact on submarine volcanic and hypabyssal rocks of the Lower Per- mian Dun Mountain ophlolite. The upper contact of the Maitai Group has not been determined because the youngest recognised Maitai strata, the Stephens Subgroup, comprising 2750 m of strata described herein, are found in syncline axes or in fault contact with younger units (Fig. 2).

The Stephens Subgroup is subdivided into five formations which are, in ascending order: the Kiwi Burn Turfs, characterised by tuff; the Acheron

ANISIAN MID TRIASSIC

SMITHIAN EARLY TRIASSIC

SANDSTONE

SILTSTONE

CONGLOMERATE

TUFF

SNOWDON FMN.

(Retford Conglomerate Member)

MASSIVE SANDSTONE ELDON SST.

RED SILTSTONE

& SANDSTONE CERBERUS FMN.

LATE PERMIAN

MASSIVE SANDSTONE ACHERON LAKES SST

TUFF & MUDSTONE KIWI BURN TUFFS

WAIUA FMN.

AGE LITHOLOGY FORMATION

Fig. 2. Summary stratigraphic column of Stephens Subgroup rocks as seen in the Countess Range (after Aitchison et al., 1988).

Dominant fithologies are shown.


Lakes Sandstones, a thick sequence of volcaniclas-

tic sandstones; the Cerberus Formation, domi-

nated by red coloured siltstones and sandstones;

the Eldon Sandstone, a sequence of volcaniclastic

sandstones; and the Snowdon Formation which

incorporates volcaniclastic sandstones, con-

glomerates, siltstones and tufts. Further details of

the Maitai Group can be found in G-rindley (1958),

Waterhouse (1964), Landis (1974, 1980), and

Aitchison et al. (1988).

Sedimentology

Facies analysis

The Stephens Subgroup is dominated by litho-

facies characteristic of subaqueous sedimentary

gravity flow deposits. Four distinctive lithofacies

which broadly accord with those of Mutti and Ricci-Lucchi (1972), Walker and Mutti (1973) and

Walker (1975) can be recognised. One additional

lithofacies comprising pyroclastic fall deposits in-

cluding tufts is also described (Table 1).

Lithofacies A

Description. Lithofacies A consists of coarse- grained sandstone and conglomerate. Beds range

from 1 to 10 m thick and generally have a lenticu-

lar geometry. Erosional basal contacts, channel-

ing, and amalgamation are common as is in-

traformational debris. Further subdivision into

three lithotypes is made on the basis of internal

structure and matrix content (Table 1). Lithofacies

A1 comprises disorganised conglomerate in which

dominantly pebble, cobble and boulder deposits

are incorporated in a generally sandy matrix.

Stratification, graded bedding, preferred clast

orientation and imbrication are lacking. Litho-

facies A2 comprises organised conglomerate which exhibits normal or inverse grading. Clasts are, on

average, cobble sized and are contained in a sandy

matrix. Lithofacies A3 comprises pebbly sand-

stones which often display massive to graded bases,

exhibit crude stratification and have coarse to

medium sand grade tops. Some beds exhibit dish

a n d / o r pillar structures and cross bedding. Exam-

TABLE 1

Lithofacies classification used in this study. General characteristics are also given

Lithofacies Characteristics (lithotype subdivision)

Sedimentary structural divisions modified after Bouma (1962) and Lowe (1982)

Reference to ~thofacies classification a

very coarse sandstones and conglomerates

(A1: disorganised conglomerate)

(A2: organised conglomerate)

(A3: pebbly sandstone)

thickly bedded coarse-medium sandstones

(BI: massive without dish structures) $3( = Ta)

(B2: massive with dish structures) $3( = Ta)

C medium to fine sandstone; shale interbeds common Ta-T e 1, 2, 3

D thin interbed of fine sandstone and shale Tb-T e 1, 2, 3

F chaotic deposits not applicable 2

T pyroclastic fall deposits: tuff not applicable

R 1

R 2, R3

S 1, 52

a References: 1 = Walker (1967); 2 =Mutti and Ricci-Lucchi (1972); 3 = Walker and Mutti (1973); 4 = modified after both Walker and Mutti (1973) and Walker (1975).

SEDIMENTOLOGY AND TECTONIC SETI'ING OF THE LATE PERMIAN-EARLY TRIASSIC STEPHENS SUBGROUP, NEW ZEALAND 59

pies of lithofacies A are exposed at Mossburn (E44 GR384950), and in Snowdon Formation beds of the Countess Range where a conspicuous conglomerate horizon can be traced over 1 km (D42 GR022557 to GR031550).

Interpretation. This lithofacies is interpreted as sediment deposited from high-density turbidity currents. Lowe (1982) discussed depositional mechanisms for these types of sediments. Very coarse gravel (A1) deposition probably occurs near instantaneously as a result of flow freezing. In- versely graded conglomerates (A2) reflect the dis- persive pressure within a traction current at the base of a high-energy, high-concentration, turbidity flow. However, normally graded units (A2) are the result of the settling out of clasts from a concentrated suspension. Lithofacies A3 is interpreted as having been deposited from a slightly unsteady but fully turbid sandy high-density turbidity current. Interaction between the flow and depositing sand bed resulted in the development of some traction structures.

Lithofacies C

Description. Lithofacies C consists of medium- to fine-grained sandstones commonly interbedded with thin mudstone (i.e. classic "proximal turbidites" of Walker, 1967). Beds are generally 1-5 m thick and typically exhibit lateral continuity. Bed contacts are generally sharp with flat erosive bed bases. Sole markings and rip-ups are common. The sand:shale ratio is typically high at around 5:1. Beds are often massive or normal graded and correspond to Bouma division Ta. Lithofacies C is common in the Cerberus Formation (D42 GR025582) and in the Snowdon Formation (e.g. D42 GR022560).

Interpretation. Features of this lithofacies suggest deposition from waning, sand-rich low-density turbidity currents.

Lithofacies D

Lithofacies B

Description. Lithofacies B consists of thickly bedded (1-2 m), moderately well sorted, medium to coarse sandstones. Beds are commonly amalga- mated resulting in a high sand:silt ratio. However, beds are more continuous and less channelised than those of lithofacies A. Intraformational debris, such as rip-ups, are common especially near bed bases. Further subdivision into two lithotypes is made on the basis of the presence or absence of water escape features such as dish or pillar structures (Table 1). Lithofacies B dominates Acheron Lakes and Eldon sandstones (D42 GR022582 and D42 GR032552) and is occasionally seen in the Snowdon Formation.

Description. Lithofacies D consists of thin interbeds of fine to very fine sandstone and siltstone (i.e. classic "distal turbidites" of Walker, 1967). It is transitional to lithofacies C. However, the lowermost Bouma T a layer is generally absent with divisions Tb_ • most common. Individual beds are generally less than 75 cm thick yet they are traceable over long (kin) distances. The sand to silt ratio is generally < 1 or << 1. Evidence of bioturbation is occasionally preserved in or on the upper surfaces of beds. Lithofacies D is seen in siltstone within the Kiwi Burn Tufts on Snowdon Peak (e.g. D42 GR057375), in Cerberus Forma- tion beds above the Mararoa River (e.g. D43 GR056269) and in the Snowdon Formation (e.g. D42 GR029550).

Interpretation. Beds of this lithofacies are interpreted as equivalent to sedimentary structural division S 3 of Lowe (1982) and are the result of rapid suspension sedimentation from high-density turbidity currents.

Interpretation. These graded sandstone beds are interpreted as deposits from low-density turbidity currents. Bioturbation is minimal which may indicate that most of the siltstone was deposited from turbidity currents rather than of hemipelagic origin.

60 J,C. A I T C H I S O N A N D C.A. L A N D I S

Lithofacies T

Description. Lithofacies T is lithologically distinctive and characterised by up to 150 cm thick beds of devitrified rhyolitic-rhyodacitic (Table 4) tuff in which silicic glass shard textures predominate. The rocks are extensively recrystallised with glass being replaced by a variety of hydrous Ca-A1 silicates including laumontite, lawsonite, prehnite and pumpellyite (Landis, 1974). Many beds are structureless and have diffuse bases whereas others have sharp bases, contain cross and parallel lamination, and convolute bedding. Tuff dominates the Kiwi Burn Tufts (Fig. 3) and is conspicuous in the Snowdon Formation.

Interpretation. Structureless tuff beds are interpreted as having formed from direct suspension sedimentation during the fallout of volcanic ash. Reworking by bottom currents a n d / o r deposition as tuff turbidites is reflected in beds which exhibit

features indicative of deposition from traction currents. These beds are physically similar to beds of lithofacies C but their distinctive mineralogy al- lows easy discrimination. Lithofacies T beds are the result of explosive volcanic activity (plinian, phreatoplinian) and are interpreted as pyroclastic fall and reworked fall deposits.

Lateral variability and cycficity

Stephens Subgroup strata exhibit degrees of lateral variability with some conglomerate horizons traceable over long distances (up to 1 km along depositional strike). Individual beds of finer-grained lithologies generally have not been traced as far. However, this probably reflects limi- tations of outcrop. Packages of distinctive lithofacies such as the Retford Conglomerate Member, Acheron Lakes sandstones and the Cerberus For- mation red coloured beds can be traced over 40 km along strike.

Fig. 3. Steeply dipping vitric tuff (lithofacies T) in the Kiwi Burn Tuffs exposed on the ridge leading east from Countess Peak. Dark-coloured beds are interbedded mudstones.

.

.-.,

~ K

iwiB

urn

Tu

rfs

' A

i~

La

klm

I

.--~

m,,

m

. |

.

~ •

~ ~

• •

~ ~

--:~,.':.,'.~:,:'~.~,~.,...'.'.",\

!:,:: ,::-.',:~i.::'.::.. ~:'~.r :.: .! ,~:::':M,=:=,','~!~:: ..~!~,, .:::,.~:~.~:.:..'..,',::.'.~,.~:.2:~,,~-,:~:-'.;:.~,~.~..~::,:

~.

~;."

"'~,~.~

,,'~s'.~.'.~i..'..'-':-:~'

c~ ~r

~.~

• ~ ~ '

.~

a.

~.~

S~

E

klo

n

San

dst

on

e I

:~.~

,:'~.~

. ~.~

.~.-..

~'.. i,.

~ ~-~

.~-,.:

~:::,-

~.,.~

'r~-~

.,~,¢-

~.~~?

,:~.~'

?~,

~-~.

,~-

0 ~

0 0

~0

O

0 o

• ~mo

wd

o~

Fo

rmat

ion

IP

o II

~"

| o

oom

oooo

¢x.

~oo

H H

-4

-4

w

~r

8 > Z t~

C~

0 t,

t"

rn

> C~

0 Z N

2:

62 J.C. A I T C H I S O N A N D C.A. L A N D I S

Fig. 5. South face of Winton Peak (Countess Range). Beds are overturned, dipping east (right). Eldon Sandstone at the top of Winton Peak with stratigraphically overlying Snowdon Formation below and to the west. Numerous fining- and thinning-upwards cycles can

be seen in Snowdon Formation.

Numerous cycles involving upsection changes in both grain size and bed thickness are a conspicuous feature of well exposed Stephens Sub- group sections in the Countess Range (Figs. 4, 5). Terminology for cycles in these subaqueous sedimentary gravity-flow sequences follows that of Ricci Lucchi (1975).

Thinning- and fining-upwards bed sequences (first-order positive cycles tens of metres thick) are frequently seen especially in the Snowdon Forma- tion (Fig. 5). Fewer thickening- and coarsening- upwards bed sequences (first-order negative cycles) are also present.

Individual Stephens Subgroup formations exhibit multiple cycles (tens to hundreds of metres thick, Fig. 4). The Acheron Lakes Sandstone beds form an overall thinning-upwards sequence (second-order positive) with an upsection trend from lithofacies B to C. The Eldon Sandstone con- stitutes an overall thickening- and coarsening-up-

ward sequence (second-order negative) which is composed of numerous smaller first-order positive cycles. The Snowdon Formation is more variable being composed of alternating positive and negative higher-order cycles.

Depositional setting

Interpretation of the depositional setting of the Stephens Subgroup is possible after detailed ex- amination of the lithofacies present and their three-dimensional architecture. A deep-marine depositional setting is indicated by both the lithofacies present and downslope transportation of marine (shelfal) macrofossils.

Cyclicity, such as that seen in Stephens Sub- group rocks can be used to infer upsection trends in the nature of sedimentation. First-order positive cycles may indicate abandonment and infill- ing of small-scale distributary channels. First-order

SEDIMENTOLOGY AND TECTONIC SE'Iq'ING OF THE LATE PERMIAN-EARLY TRIASSIC STEPHENS SUBGROUP, NEW ZEALAND 63

negative cycles indicate the progradation of small depositional lobes. Multiple cycles indicate similar processes on a larger scale. Overall epiclastic sedimentation in the Stephens Subgroup is dominated by numerous small multiple positive cycles. A complex pattern is present and it suggests migration of small-scale distributary channels and depositional lobes during Stephens Subgroup accumulation. The presence of such variable hthology and numerous cycles within the Stephens indicate proximity to migratory feeder channels somewhere on an active part of a submarine fan or ramp complex. In an active island arc-related environ- ment migration or switching of feeder channels is likely to have been the result of intrabasinal tectonics.

Large-scale cychcity within sedimentary gravity-flow sequences has been used widely in en- vironmental interpretation (Ricci Lucchi, 1975; Walker and Mutti, 1973; Normark, 1978). Clearly there is a complex interplay of many variables which control sedimentation patterns on submarine fans and ramps (Nelson and Nilsen, 1984; Heller and Dickinson, 1985). Presently we do not have sufficient data to distinguish exactly where Stephens sedimentation took place and we share the concern which has recently been expressed about the widespread use of detailed environmen- tal interpretations of ancient submarine fan deposits (Shanmugam et al., 1985). Although comparison with examples of both modern and ancient deep-sea fans (Nelson and Nilsen, 1984) suggests that Stephens Subgroup sediments may have been deposited in a mid-fan area we note that Stephens Subgroup depositional cycles are

,generally smaller in scale than those which would be expected in many submarine fan environments.

Maitai Group exposure is inferred to parallel depositional strike (Landis, 1980). The absence of evidence for point sources or input from major rivers, no evidence of large-scale channels together with widely developed structureless and amalga- mated sand bodies, which commonly show water expulsion structures, suggests rapid deposition of epiclastic sediment cascading into deep water from shelf depths without the development of radial distributary channels on fan-shaped aprons (Chan and Dott, 1983).

The hthofacies present and the regional extent of lithofacies packages indicates deposition on a submarine fan or mass flow apron complex proximal to an active volcanic arc. Small-scale fans may have developed at the base of an extensive slope apron in a manner :analogous to the Creta- ceous backarc rocks described from Antarctica (Ineson, 1989). Contemporaneous explosive volcanic episodes are also evident from Stephens strata, with numerous felsic tuff beds comprising much of the Kiwi Burn Tufts and also forming conspicuous marker horizons within the Snowdon Formation.

Duration of Stephens Subgroup sedimentation

The dominance of coarse-grained lithofacies suggests proximity to sediment sources and rapid deposition. Sedimentation rates in hthologically similar sediments (Atka Basin, Aleutian Ridge and Komandorskiy Basin; Stewart, 1977, 1978) are in the order of 300 m Ma -1 for unhthified sediments. Ingersoll (1979) indicates a similar rate (265 m Ma -1) for indurated Great Valley Se- quence strata of a similar lithofacies association and we estimate an overall rate of 260 m Ma- 1 for sedimentation of the volcaniclastic Triassic strata of the Taringatura Hills, Southland (Coombs, 1950). If a comparable rate (300 m Ma -1) is assumed for the Stephens Subgroup then the ap- proximate time for accumulation would be in the order of 9 Ma.

Sandstone petrography

Stephens rocks are dominated by first-cycle epiclastic volcanogenic sandstones and associated mudstones. Typical sandstones contain less than 2% clastic quartz. Lesser amounts of vitric tuff and tuffaceous sandstone are also present. The Stephens has experienced varying degrees of burial metamorphism. However original texture and mineralogy can usually be recognised.

Sandstone modal analyses, combined with whole rock chemical analyses and microprobe-determined compositions of detrital pyroxenes are presented here. These place constraints on sediment provenance.


Samples were collected from throughout the studied area and, following reconnaissance petrographic studies, the least altered were selected for detailed analysis. Even the least altered samples have experienced thorough albitization of plagioclase, widespread chloritization of mafic minerals, and growth of authigenic cements. Nev- ertheless, it is usually possible to reconstruct original clast compositions and framework-matrix relations.

Matrix content of the sandstones is highly variable and its origin is frequently problematic (Dickinson, 1970). We follow the procedures of Mackinnon (1980, 1983), rejecting samples con- taining greater than 20% matrix. Counting procedures follow those proposed by Dickinson (1970), Ingersoll (1978) and others, thereby facilitating comparison with other described suites. 400 points were counted for each sample using a stage inter- val greater than the diameter of the largest grain. Thus statistical error due to variation in sample population should not exceed 5% at the 90% confi- dence level (Van der Plas and Tobi, 1965). Micro- phenocrysts larger than 0.0625 mm are counted using the GaTxi Dickinson methodology (Ingersoll et al., 1984) and are placed in the appropriate mineral category, rather than treated as lithic components.

A selection of 10 samples were stained for plagioclase and potassic feldspar. No potassic feldspar was found and the distinction between quartz and plagioclase was found to be quite obvious. As a result further staining was deemed unnecessary.

Detrital components

Point counting categories and results are given in Table 2. The major detrital components are volcanic lithic fragments and plagioclase.

Quartz grains (monocrystalline) are characterised by unit extinction, straight or embayed grain boundaries and the absence of inclusions. Quartz is of volcanic and amygdaloidal origin. Rare polycrystalline quartz contains vacuoles and shows semi-composite to undulose extinction; it is probably derived from the erosion of quartz veins.

Plagioclase is readily distinguished from quartz by the abundance of inclusions, contrasting relief, and extensive twinning. Although albitized, the originally calcic nature is evident from the widespread fine-grained Ca-A1 silicate inclusions and cements (mainly prehnite and pumpellyite; minor lawsonite and laumontite). Phyllosilicate inclusions, particularly chlorite and sericite are also abundant.

Clinopyroxene is found widely, but seldom seen in abundance. It occurs as large coloudess crystals (/3 = 1.705, 2~ = 56 °, zAc = 57°). Most grains show some degree of alteration, particularly with marginal replacement by chlorite and filling of internal dissolution cavities by quartz. Prehnite and pumpellyite replacement is also recorded. No other mafic minerals are recorded from Stephens sandstones. However, pseudomorph habits combined with bulk rock chemical compositions suggests that both orthopyroxene and olivine may have been present originally. There is no evidence that either hornblende or biotite were ever present in any abundance.

Minor clastic mineral components include mag- netite, carbonaceous material (plant scraps), calcite (fossil fragments), epidote and apatite.

Lithic volcanic clasts are the predominant com- ponent in most Stephens sandstones. A range of igneous textures is recognised, but in most cases, alteration precludes detailed analysis of textural types. Two broad categories are recognised: (a) vitric plus felsitic, and (b) microlitic plus lath- work. Predominance of the latter category indicates a mainly basaltic to andesitic provenance.

Lithic sedimentary grains (sandstone and siltstone) are an uncommon but widely present com- ponent. All closely resemble the associated strati- fied rocks. They are regarded as being of intraformational origin and although included in point counts (Table 2) are excluded from recalcu- lation for QFL diagrams (Fig. 6).

Matrix includes both recrystallised "true" matrix of fine-grained clastic detritus (orthomatrix of Dickinson, 1970) and diagenetic pore filling and grain-replacing matrix (epimatrix of Dickinson, 1970). Squashed lithic grains (pseudomatrix of Dickinson, 1970) are classified where possible into their respective lithic categories and not included

TA

BL

E 2

Mea

n an

d 1

s.d.

res

ults

of

sand

ston

e de

trit

al c

ompo

siti

on a

naly

ses

of S

teph

ens

Sub

grou

p sa

mpl

es

Loc

alit

y C

ount

ess

Ran

ge-S

now

dow

n P

eak-

Mar

aroa

Riv

er

Nor

ther

n co

rrel

ativ

es

Lit

holo

gica

l un

it

Ste

phen

s S

ubgr

oup

Ach

eron

Lak

es

Cer

beru

s E

ldon

S

now

don

Sno

wdo

n av

erag

e S

ands

tone

F

orm

atio

n S

ands

tone

F

orm

atio

n a

For

mat

ion

b

Mea

n 1

s.d.

ra

nge

mea

n 1

s.d.

m

ean

1 s.

d.

mea

n 1

s.d.

m

ean

1 s.

d.

mea

n 1

s.d.

Ste

phen

s S

ubgr

oup

aver

age

mea

n 1

s.d.

ra

nge

Qua

rtz

(mon

oery

stal

line

) 1.

4 (1

.3)

0-

6 1.

5 (1

.5)

3.1

(0.2

) 1.

1 (0

.7)

1.7

(1.4

) 0.

7 (0

.5)

1.0

(0.9

) 0

- 3

Qua

rtz

(pol

ycry

stal

line

) 0.

4

Fel

dspa

r 20

.7

Lit

hic

vole

. (fe

lsit

ic)

12.3

L

ithi

c vo

le. (

mie

roli

tic)

46

.7

Lit

hic

sed.

(si

lt)

2.7

Lit

hic

sed.

(sa

nd)

0.9

Pyr

oxen

e 1.

3 M

isce

llan

eous

7.

7

Mat

rix

6.3

Qua

rtz

(tot

al)

2.2

Fel

dspa

r 26

.1

Lit

hic

(tot

al)

71.6

Num

ber

of s

ampl

es

70

(0.5

) 0

- 2

0.5

(0.5

) 1.

1 (0

.4)

0.4

(0.3

) 0.

5 (0

.6)

0.2

(8.5

) 7-

45

26.2

(7

.4)

23.9

(0

.1)

23.1

(6

.5)

17.3

(6

.6)

22.2

(7

.1)

0-44

14

.7

(10.

2)

15.0

(5

.7)

11.1

(4

.9)

13.8

(6

.8)

5.8

(9.2

) 15

-67

4.4

(9.4

) 50

.5

(5.0

) 49

.6

(6.3

) 44

.9

(9.6

) 51

.5

(1.2

5)

0-21

0.

8 (1

.3)

- -

1.1

(1.9

) 0.

8 (1

.1)

1.6

(0.3

5)

0-

6 0.

1 (0

.1)

- -

0.2

(0.3

) 0.

6 (1

.3)

0.8

(2.7

) 0-

16

1.4

(2.8

) 1.

0 (0

.4)

1.1

(1.9

) 1.

9 (3

.9)

1.3

(7.0

) 1-

31

7.5

(8.1

) 4.

1 (1

.9)

5.3

(2.9

) 1.

9 (7

.5)

7.1

(4.4

) 0-

15

3.5

(2.1

) 1.

4 (1

.4)

5.8

(4.1

) 8.

2 (4

.8)

9.0

(1.9

) 2

4 1

2 1

(8.5

) 30

26

27

22

27

(8

.0)

68

70

72

76

72

10

2 10

23

5

(0.3

) (4

.7)

(4.6

) (4

.3)

(1.2

) (1

.1)

(1.0

) (1

.2)

(2.7

)

0.4

(0.5

) 24

.8

(8.5

) 7.

6 (6

.0)

52.4

(9

.2)

0.4

(1.0

) 0.

3 (0

.7)

0.3

(0.6

) 3.

7 (2

.0)

6.1

(4.6

)

1.4

(0.9

) 28

.5

(8.9

) 70

.1

(8.6

)

16

0-2

4-

37

2-21

31

-69

0-4

0-3

0

-3

1-7

0-

18

a JC

A s

ampl

es.

b O

ther

sam

ples

.

66

T A B L E 3

Resu l t s o f m i c r o p r o b e ana lyses o f S t ephens S u b g r o u p de t r i t a l p y r o x e n e c o m p o s i t i o n s

J.C. AITCHISON AND C.A. LANDIS

Specimen 0U49604

SiO 2 52.42 48 .70 49 .67 52.37 51.32 52.36 52.89 51.05 50.38 51.82

A1203 2.44 10.87 5.40 1.86 1.90 3.42 2.82 2.32 3.60 2.60

T i O 2 0.46 0.22 0.51 0.40 0.61 0.18 0.36 0.42 0.31 0.61

F e O * 4.64 10.90 10.63 10.24 10.53 5.00 6.84 8.35 6.45 7.62

M n O 0.01 - 0.05 0.06 - - - 0.08 - 0.05

M g O 18.74 9.46 14.92 15.36 15.24 17.86 16.57 15.91 16.83 15.85

C a O 24.94 20.46 18.62 19.63 19.87 20.83 21.07 20.71 21.33 20.93

N a 2 0 0.19 0 .19 0.24 0.33 0.33 0.16 0.23 0.34 0.26 0.27

K 2 ° 0 .02 0.02 0.09 0.02 0.01 0.01 - 0.01 0.01 0.01

Cr203 0.36 - - - 0,06 0.31 - 0.02 0 .40 0.05

T o t a l 100.91 100.82 100.12 100.28 99,85 100.15 100.77 99.19 99 .57 99 .80

Specimen 0U49610

SiO 2 51.42 51.46 51.55 50.66 51.76 51.06 51.71 51.81 52.11 50.77

A1203 2.41 2.74 2.94 2.83 1.99 2.61 1.68 2.32 1.24 4.29

T i O 2 0.55 0.77 0 .44 0.69 0.56 0.33 0.60 0.51 0 .54 0.93

F e O * 8.04 8.60 8.99 8.85 6.94 5.07 10.45 7.96 11.33 8.47

M n O - 0.06 0.02 0.05 0.08 - - 0 .03 0.02

M g O 16.17 15.78 16.20 16.04 18.10 17.49 14.70 16.57 15.10 13.55

C a O 21.11 20.42 19.68 20.94 19.57 22.13 19.96 20.26 19.51 21.42

N a 2 0 0.37 0.30 0 .30 0.33 0.31 0.21 0.36 0.30 0.31 0 .34

K 2 ° 0.01 - 0.01 0.02 0.02 0.01 - 0.02 0.01 -

Cr203 0.05 0.03 0.22 0.13 0.38 0.54 0 .04 0.13 - -

To ta l 100.12 100.15 100.35 100.51 99.70 99.45 99.50 99.86 100.16 99.79

Specimen 0U49639

SiO 2 52.74 54.01 51.91 49.65 52.65 51.74 51 .50 52.38 49 .80 50.28

A I 2 0 3 0.87 1.18 1.65 5.15 1.17 2.49 3.18 2.23 4 .72 5.00

T i O 2 0.12 0.11 0.39 1.13 0.26 0.67 0.87 0.53 0 .54 0.55

F e O * 8.53 3.88 8.78 8.29 8.77 7.63 7.27 6.53 9.51 6.00

M n O 0.03 - - - 0 .06 - 0 .10 - - 0 .02

M g O 15.43 18.50 15.27 15.25 15.23 16.26 16.22 17.46 15.99 15.85

C a O 22.00 21.42 21 .04 20.85 21.14 20.28 20.33 21.11 18.93 22.23

N a 2 0 0.41 0.27 0.38 0.28 0.36 0.31 0.38 0.25 0.31 0.33

K 2 ° 0.01 - 0.01 - 0.01 0.03 0.01 0.03 0.06 0 .02

C r 2 0 - 0.76 - 0.25 0.08 - - 0.23 0.11 0.20

To ta l 100.14 100.14 99 .42 100.85 99.72 99.41 99.85 100.75 99.95 100.50

Specimen 0U49616

SiO 2 52.46 51.41 54.03 52.15 48.37 52.53 53.83 52.87 51 .84 52.70

A1203 1.41 2.35 0 .80 1.39 3.08 1.41 1.25 1.10 1.44 1.15

T i O 2 0.25 0.43 0.16 0 .30 0.27 0.23 0.18 0.13 0.28 0.13

F e O * 8.64 8.47 8.44 8.72 10.93 8.50 4.28 3.72 8.69 8.24

M n O 0.07 0.08 - 0.05 - - - 0 .07 - 0 .04

M g O 15.17 14.60 13.84 15.27 14.08 15.04 16.83 18.49 14.89 14.72

C a O 21.25 22.42 22.53 21.89 22.63 21.99 22.75 22.97 21.98 21 .92

N a 2 0 0.44 0,36 0 .30 0.33 0.32 0.40 0.23 0 .24 0,41 0.38

K 2 ° - 0.01 0.01 0.01 0 .04 0.01 0.01 0.02 0,01 -

Cr203 - - 0 .03 0.02 0.05 0.05 0.57 0.51 0 .12 0 .04

T o t a l 99.67 100.12 100.13 100.22 99.77 100.16 99.91 100.12 99 .66 99.31

SEDIMENTOLOOY AND TECTONIC SETFING OF THE LATE PERMIAN-EARLY TRIASSIC STEPHENS SUBGROUP, NEW ZEALAND

TABLE 3 (continued)

67

Specimen Stephens Subgroup from Mossburn

SiO 2 50.62 51.43 51.21 50.98 51.01 50.70 51.50 51.93

AI20 a 2.06 2.50 1.98 1.96 2.39 2.72 2.58 1.79

TiO 2 0.37 0.44 0.39 0.54 0.37 0.33 0.35 0.41

FeO * 12.08 11.57 11.91 14.16 11.98 11.84 11.75 13.33

MnO - 0.03 0.06 0.01 0.05 - - -

MgO 16.56 16.59 16.66 14.70 16.78 16.47 16.55 15.62

CaO 17.22 16.84 17.31 17.14 17.10 18.06 17.01 15.99

N a 2 0 0.25 0.26 0.20 0.31 0.28 0.26 0.26 0.18

K 20 - - 0.01 0.01 0.01 0.01 - 0.01

Cr203 - 0.03 0.04 - 0.03 0.05 0.13 -

Total 99.17 99.68 99.76 99.84 99.99 100.44 100.13 100.45

Specimen Acheron Lakes Sandstone from Winton Peak 0U49639

SiO 2 50.82 51.69 49.96 49.61 52.26 51.08 51.36 50.64

AI203 1.69 3.25 4.40 2.31 1.41 3.24 3,41 4.11

TiO 2 0.48 0.77 0.81 0.68 0.28 0.83 0,76 0.85

FeO * 9.54 7.36 7.42 11.26 8.94 9.44 9,25 6.06

MnO - - 0.02 - 0.02 - - -

M 8 0 16.23 15.57 15.84 16.17 14.96 14.90 16.04 15.67

CaO 21.63 21.63 20.85 19.12 21.54 19.85 19.22 22.70

N a 2 0 0.34 0.26 0.35 0.33 0.42 0.45 0,48 0.42

K 2 0 - 0.01 0.02 0.01 0.01 0.04 - -

Cr203 0.12 0.03 0.04 0.05 0.07 0.06 - 0.21

Total 00.8i 100.56 99.70 99.55 99.89 99.88 100,51 100.64

within the matrix total. Epimatrix predominates and typically includes quartz, prehnite, pumpellyite, laumontite, chlorite, calcite, sericite and titanite.

Results

Detrital (Q : F : L) modes of modem sands and ancient sandstones vary according to provenance

OLin Lv O O O

F L L L L

Fig. 6. Stephens Subgroup sandstone detri tal Q : F : L. composition plots Countess Range, Snowdon Peak, Mararoa River and Nelson

samples. Diagram at the left shows mean and 1 s.d. for all northern Southland samples.

68 J.¢. AITCHISON AND C.A. LANDIS

types as governed by tectonic setting (Dickinson and Suczek, 1979; Dickinson et al., 1983). Stephens sandstones [Q : F : L = 2 : 26 : 72] are typical of magmatic arc provenance (Dickinson and Suczek, 1979) being comparable to those of both back arc (West Philippine Basin [Q : F : L = 5 : 17 : 78], Har- rold and Moore, 1975) and some fore arc regions (Middle America Trench [Q : F : L = 1 : 43 : 56], Enkeboll, 1981; [Q : F : L = 3 : 21 : 76], Yerino and Maynard, 1984; and Atka Basin, Alaska [Q : F : L = 6 : 33 : 60], Stewart, 1978).

The high proportion of volcanic lithic and plagioclase grains and corresponding rarity or absence of detrital quartz, K-feldspar and metamorphic lithic grains indicates a volcanic island arc as opposed to a continental arc while the high proportion of lithic clasts relative to feldspar and quartz is characteristic of a relatively undissected volcanic island arc.

Detri tal pyroxene composit ion

Clinopyroxene occurs as a detrital phase in many Stephens rocks, particularly the basaltic and andesitic sandstones. 56 grains from 6 widely spaced samples were analysed using the electron microprobe analyser. Results are presented in Ta- ble 3 and summarised graphically in Figs. 7 and 8. These detrital grains consist mainly of augite, but also include diopside, endiopside and salite (Fig. 7). They compare closely in composition with volcanic arc basalt and basaltic andesite clinopyroxene as summarised by Nisbet and Pearce (1977) and lie mainly within Le Bas' (1962) field of clinopyroxene from sub-alkaline volcanic suites. In particular on a discriminant analysis plot (Fig.

Diopside

/ ~, ./~. S.,.e / .". "~.':'i!. • "Aug|t e

Mg Fe+Mn

Fig. 7. Ca-Mg-Fe plot showing composition of Stephcns Subgroup sandstone detrital pyroxenes.

-1~ -~o

WPA

F1 - 0 .,9 - 0 .ill

! i i I VAB

/ • t °

~ " OFB

WPT ~ OFB ~ .

WPT

--2.4

F2

~-2.5

~-2.6

,2,7

Fig. 8. Discriminant functions F 2 vs. F 1 of Stephens Subgroup sandstone detrital pyroxenes (after Nisbet and Pearce, 1977). VAB = volcanic arc basalt, OFB = ocean floor basalt, WPA =

within plate alkali, WPT = within plate tholeiitic.

8) proposed by Nisbet and Pearce (1977) the analyses plot entirely within the volcanic arc basalt and basaltic andesite fields.

Whole-rock geochemis try

Due to the immature first-cycle nature of Stephens volcanic sandstones, their chemical compositions give good approximations to the nature of the eroding source area. Analyses of interbedded tufts provides information on the nature of c o n t e m p o r a n e o u s explosive volcanism. 16 Stephens rocks (10 sandstones, 6 tufts) were analysed using XRF methods (Table 4). Although the sandstones are mineralogically reconstituted, we believe that except for possible sodium en- hancement, their whole-rock chemistry is largely unaltered: authigenic minerals are very fine- grained, delicate sedimentary structures are well preserved and undeformed, and veins are generally absent. Recrystallisation is further advanced in the tufts with post-depositional compositional mottling well developed (Landis, 1974; Fig. 5). A study of similar Triassic tufts in southern New Zealand (Boles and Coombs, 1975) reveals consid- erable element mobility. Garcia (1978) showed

S E D I M E N T O L O G Y A N D T E C T O N I C S E ' I ' F I N G O F T H E L A T E P E R M I A N - E A R L Y T R I A S S I C S T E P H E N S S U B G R O U P , N E W Z E A L A N D 69

5.00

1) 4 .00

,~ 3.00 IK o i,• 2 .00

1.00

Tholeiitlc field ~ ~ o • e e ~ e o

e C a l c _ a l k a l i n e f i e ld °

60.0 70.0 80.0 SJO 2

Fig. 9. FeO*/MgO vs. SiO 2 plot of Stephens Subgroup sandstones (solid circles) and tuff (open circles) analyses. Tholeiitic

and cale-alkaline fields after Garcia (1978).

that certain elements (particularly trace elements) remain relatively immobile during alteration of arc-volcanic rocks. Thus as a check on remobiliza- tion, we have included selected trace elements in our analyses. Fig. 10 confirms that titanium and zirconium in sandstones are not likely to have been significantly altered; in contrast the wide scatter of points from tufts suggests that even these components may experience mobility in originally glassy sediments. Extreme variations in sodium and rubidium (Table 4) indicate mobility of at least some components during diagenesis and

5JQ

T ~ (%)

1.0

0.5

0.1 10

/ " ' " " " f i e l d ,,

/

; . 'ORB ;

",, / . " $1eld / , '

4 J s. / t .s "p e ee " \ /s ; • i s " ",

, 'Arc f ield " ' ~ ' " " it

i i i i

| i i , i , | | H L

X k

k \

\ X

\ x

%

i

~t t

0

i i J i i I i | H

Zr Ipl~ml 1000

Fig. 10. Ti vs. Zr plot of Stephens Subgroup sandstones (solid circles) and tuff (open circles). Compositional fields for pre~ sent-day volcanic rocks from island arc, ocean ridge basalts

(ORB) and within-plate settings are after Pearce (1980).

low-grade metamorphism of these tufts (see also Boles and Coombs, 1975).

Petrographic data as well as chemical analyses of detrital clinopyroxene (previous sections) have already been used to argue for a volcanic arc source. The new whole-rock analyses, when com- pared with various volcanic and sedimentary rock chemographic plots, conclusively support arc provenance (see also Roser and Korsch, 1988). Sandstones are andesitic, while tufts are dacitic to rhyolitic in character. Anhydrous SiO 2 ranges between 54 and 62% for sandstones and between 69 and 74% for tufts. Both tholeiitic and calc-alkaline affinities are seen in major element plots (Fig. 9). A plot of the relatively immobile elements Ti vs. Zr (Garcia, 1978; Fig. 10) shows distinctly calc-alkaline character for both epiclastics and pyroclas- tics.

T e c t o n i c se t t ing

Stephens Subgroup rocks form part of the fault-bounded Dun Mountain-Maitai terrane (Fig. 1; Bishop et al., 1985). This terrane consists of a thick synclinally folded or westward younging pile of deep-water volcanidastic sediments and limestones deposited on an ophiolitic basement. The ophiolitic basement, The Dun Mountain Ophiolite Belt (DMOB), is extensively disrupted with m61ange present along the regionally extensive Livingstone fault system where it lies in contact with highly deformed, sparsely fossiliferous late Paleozoic-Mesozoic rocks which represent the adjoining Caples terrane. Lowermost Maitai Group sediments consist of polymict breccias which con- formably and locally disconformably overlie the ophiolitic sequence from which they are in part derived; overlying Maitai Group rocks include bioclastic limestones and volcanic arc-derived epiclastics including the Stephens Subgroup. The Dun Mountain-Maitai terrane is fault bounded to the west along the Hollyford fault system against relatively fossiliferous volcanogenic sediments of the Triassic-Jurassic Mudhiku terrane, or Cenozoic strata.

Several previous interpretations of the terrane (e.g. Coombs et al., 1976; Davis et al., 1980; Mackinnon, 1983) have depicted the Dun Moun-


tam-Maitai terrane developing adjacent to the terranes which presently bound it. These interpretations have generally supported a model of westward dipping subduction (cf. Howell, 1980) with the Brook Street terrane representing a volcanic island arc, Maitai Group and the Murihiku terrane representing forearc basins and the Caples terrane, in the east, representing an associated accretionary prism.

Development of the terrane concept (Coney et al., 1980; Jones et al., 1983) and increased under- standing of tectonic processes require that unless

(or until) adjacent terranes can clearly be shown to be related (and therefore composite terranes), such a relationship should not be assumed. Whole-rock geochemistry argues against deriva- tion of Maitai and Murihiku sediments from a Brook Street source (Landis and Blake, 1987). Although Murihiku terrane epiclastics are similar to those of the Dun Mountain-Maitai terrane, metamorphic studies near Lintley and Clinton (Cawood, 1986, 1987) show that the very tow grade of Maitai Group rocks is inconsistent with the hypothesis of their burial by a pile of sedi-

TABLE 4

Results of whole-rock geochemical analyses of Stephens Subgroup sandstones and tufts

OU49601 OU49605 OU49612 OU49641 OU49654 OU49655 OU49675 OU49686

sandstone sandstone sandstone sandstone sandstone sandstone sandstone sandstone

Major elements (96)

SiO 2 51.69

A1203 17.19

FeO(tot) 9.60

MnO 0.18

MgO 3.25

CaO 5.49

Na 20 4.68

K20 1.34

v~o~ 0.30 TiO 2 1.14

LOI 4.75

Total 99.61

Trace elements (ppm)

Ba 246

Ce 33

Cr 49

Cu 57

Ga 21

La 10

Nb 3.7

Nd 19

Ni 22

Pb 15

Rb 39

Sc 27.4

Sr 168

Th 4.6

U 1.4

V 235

Y 32

Zn 105

Zr 138

57.88 54.89 54.24 52.59 55.03 56.74 60.32

16.53 17.17 17.00 17.77 16.79 16.07 16.32

7.40 10.01 7.92 8.40 8.27 8.85 6.66

0.11 0.13 0.12 0.12 0.12 0.14 0.10

3.01 3.27 4.36 4.52 4.08 3.89 2.56

4.90 2.91 7.17 7.25 6.85 3.92 4.52

4.07 4.87 3.53 3.58 2.77 4.69 3.82

1.16 0.89 0.05 0.22 0.59 0.95 1.67

0.17 0.22 0.15 0.32 0.20 0.20 0.18

0.94 1.19 0.95 0.95 0.87 1.11 0.87

3.80 4.60 4.96 4.50 4.58 3.53 3.20

99.97 99.85 100.45 100.22 100.15 100.09 100.22

250 183 102 87 142 211 205

31 34 25 19 21 36 39

69 46 76 79 48 63 57

37 63 41 62 32 37 30

19 19 23 20 20 19 22

11 8 7 4 9 15 18

5.5 3.6 3.7 3.7 1.0 4.4 6.1

21 21 14 10 13 15 29

24 23 30 40 20 26 23

13 12 14 13 10 14 17

31 23 3 7 13 26 44

20.7 24.8 20.2 24.9 22.8 21.8 19.1

155 271 146 523 186 152 106

5.9 4.8 3.6 1.0 4.5 6.0 7.7

1.3 1.1 1.2 1.1 1.3 1.4 1.7

165 245 199 215 232 186 137

26 33 18 21 20 30 32

91 110 92 91 98 105 89

156 140 126 112 114 171 187

SEDIMENTOLOGY AND TECTONIC SETTING OF T H E LATE P E R M I A N - E A R L Y TRIASSIC STEPHENS SUBGROUP, NEW ZEALAND

TABLE 4 (continued)

71

OU49684 O U 4 9 6 8 5 OU49611 O U 4 9 6 1 7 O U 4 9 6 1 9 O U 4 9 6 3 3 O U 4 9 6 8 3 OU49634 sandstone sandstone tuff tuff tuff tuff tuff tuff

Major elements (%) SiO 2 54.03 55.33 64.36 67.72 67.88 66.78 A1203 16.29 17.80 14.08 12.94 13.41 16.76 FeO(tot) 8.41 8.24 3.55 2.92 3.87 3.24 MnO 0.14 0.13 0.06 0.02 0.09 0.04 MgO 4.24 2.68 0.90 0.89 1.61 0.75 CaO 7.42 3.78 6.82 6.06 4.57 3.19 Na 20 3.23 5.49 0.03 0.01 1.25 1.53 K20 0.37 0.95 0.20 0.35 1.57 2.93 P205 0.17 0.18 0.09 0.09 0.11 0.05 TiO 2 0.98 0.99 0.35 0.26 0.40 0.27 LOI 4.89 4.42 9.78 9.29 5.49 4.43

Total 100.17 99.99 100.22 100.55 100.25 99.97

Trace elements (ppm) Ba 84 216 94 108 336 315 763 259 Ce 24 27 37 42 54 50 89 38 Cr 96 52 10 6 6 10 7 3 Cu 41 49 16 10 14 17 10 3 Ga 20 21 18 18 17 20 27 15 La 11 12 15 22 21 18 42 14 Nb 3.9 3.1 6.9 6.3 10.2 9.0 13.6 9.4 Nd 13 20 16 20 32 27 46 22 Ni 34 24 7 4 7 8 7 3 Pb 13 14 23 30 20 23 26 23 Rb 14 34 9 9 78 45 92 26 Sc 24.6 21 7.6 6.5 10.6 13.3 8.9 4.6 Sr 162 177 656 538 204 162 131 246 Th 3.0 3.9 17.7 20.9 14.1 15.7 22.6 11.8 U 1.2 1.4 3.1 3.1 3.1 2.6 3.5 2.1 V 216 181 42 29 39 52 36 13 Y 21 24 26 18 34 51 50 27 Zn 93 101 46 40 76 86 96 47 Zr 124 127 160 142 303 267 339 101

ments as thick as those of the Mur ih iku terrane.

The Caples terrane (TurnbuU, 1979a, b, 1980),

s i tuated to the east of D M O B melange rocks,

includes strata in terpre ted as having deve loped in

an accret ionary prism. However , the Caples ter-

rane is largely unfossi l i ferous and of uncer ta in late

Pa l e ozo i c -Mesozo i c age. Det r i ta l sandstone com-

posi t ions (TurnbuU, 1979b; Mackinnon , 1983) are

significantly more quar tzose than those of the

D u n M o u n t a i n - M a i t a i terrane and whole rock

and n e o d y m i u m geochemis t ry of Caples ter rane

sandstones is incompat ib le wi th der ivat ion f rom a

Brook Street source (Landis and Blake, 1987; Fros t

and Coombs , 1987).

A l though detai ls of the evolu t ionary his tory of

the D u n M o u n t a i n - M a i t a i terrane remain un-

known, field sedimentology, detr i tal sandstone

pe t rography, minera logy and chemis t ry clearly in-

dicate deve lopmen t in a deep-sea sett ing in close

p rox imi ty to an act ive oceanic volcanic is land arc.

A lack of cher t or o ther pelagic sediments be tween

the ophiol i te and over ly ing sediments suggests

Mai ta i sed iment was no t deposi ted upon older

oceanic crust in a mid ocean or t rench sett ing and


we beheve a more likely depositional setting is forearc or backarc basin. In the absence of either an arc source terrane and associated paleocurrent data, it is impossible to resolve whether the Maitai accumulated in a forearc or backarc region. If this terrane did accumulate in a forearc region then the presence of ophiolitic rocks is of particular importance.

Forearc sediments on ophiohtic basement have been described by several workers (e.g. Ingersoll, 1979; Hussong et al., 1982; Lundberg, 1982; For- sythe et al., 1986). However, in all the modem forearcs flanking oceanic island arcs with which we are famihar, there is no evidence for development of a continuous trough floored by contemporaneous ophiohte extending more than 300 krn in length. Similarly, there is no evidence for regionally extensive clastic stratigraphic units within modem oceanic forearc basins. In fact, our im- pression is that most forearc basins flanking oceanic arcs tend to be very complex and locally highly variable in terms of both basement geology and sediment cover. Back arc basins, on the other hand, have greater lateral continuity and are un- derlain by oceanic crust of ophiohtic character (see also Busby-Spera, 1988).

After its development the Dun Mounta in- Maitai terrane was tectonically removed from strata which developed adjacent to it. Dun Moun- tain-Maitai terrane and terranes which now bound it have undergone an as yet undetermined degree of tectonic transportation relative to each other.

Conclusions

(1) The nature of sedimentary lithofacies within the Stephens Subgroup and the dominance of regionally extensive units indicates a probable apron like source opposed to the development of these strata as a point-source fan.

(2) Sedimentary petrography and whole-rock geochemistry clearly indicate calc-alkaline oceanic island arc provenance for Stephens Subgroup sediments.

(3) The absence of any clearly recognisable source terrane suggests that the Dun Mounta in- Maitai terrane is not genetically related to terranes against which it is presently juxtaposed.

Acknowledgements

This manuscript has benefitted from comments by R.A.F. Cas, D.S. Coombs, D.D.L. Pillai and the reviewers. We are most grateful to B. Roser for geochemical analyses. Financial support provided by a University of Otago research grant is also acknowledged.

References

Aitchison, J.C., 1984. Stephens Subgroup (Upper Maitai Group) in the Countess Range-Mararoa River area. M.Sc. Thesis. University of Otago, Dunedin, 165 pp.

Aitchison, J.C., Landis, C.A. and Turnbull, I.M., 1988. Stratig- raphy of Stephens Subgroup (Maital Group) in the Countess Range-Mararoa River area, northwestern Southland, New Zealand. J. R. Soc. N.Z., 18: 271-284.

Bishop, D.G., Bradshaw, J.D. and Landis, C.A., 1985. Provi- sional Terrane Map of South Island, New Zealand. In: D.G. Howell (Editor), Tectonostratigraphlc terranes of the Circum-Pacific Region. Circum-Pac. Counc. Energy Miner. Resour., Earth Sci. Sen, 1: 515-521.

Boles, J.R. and Coombs, D.S., 1975. Mineral reactions in zeolitic Triassic tuff, Hokonni Hills, New Zealand. Geol. SOc. Am. Bull., 86: 163-173.

Busby-Spera, C.J., 1988. Evolution of a Middle Jurassic backarc basin, Cedros Island, Baja, California. Geol. Soc. Am. Bull., 100: 218-233.

Campbell, H.J., Coleman, A.C., Johnston, M.R. and Landis, C.A., 1984. Geology of Stephens Island and the age of Stephens Formation. N.Z.J. Geol. Geophys., 27: 277-289.

Cawood, P.A., 1986. Stratigraphic and structural relations of the Dun Mountain Ophiollte Belt and enclosing strata, northwestern Southland, New Zealand. N.Z.J. Geol. Geo-

phys., 29: 179-204. Cawood, P.A., 1987. Stratigraphic and structural relations of

the strata enclosing the Dun Mountain Ophiolite Belt in the Arthurton-Clinton region, Southland, N.Z.J . Geol.

Geophys., 30: 19-36. Chart, M.A. and Dott, R.H.J., 1983. Shelf and deep-sea sedi-

mentation in an Eocene forearc basin, Western Oregon-- fan or non-fan? Bull. Am. Assoc. Pet. Geol., 67: 2100-2116.

Coney, P.J., Jones, D.L and Monger, J.W.H., 1980. Cordil- leran suspect terranes. Nature, 288: 329-333.

Coombs, D.S., 1950. The geology of the Northern Taringatura Hills. Trans. R. SOc. N.Z., 82: 65-109.

Coombs, D.S., Landis, C.A., Norris, R.J., Sinton, J.M., Borns, D.J. and Craw, D., 1976. The Dun Mountain ophiolite belt, New Zealand, its tectonic setting, constitution, and origin, with special reference to the southern portion. Am. J. Sci.,

276: 561-603. Davis, T.E., Johnstone, M.R., Rankin, P.C. and Stall, R.J.,

1980. The Dun Mountain ophiolite belt in east Nelson,

SEDIMENTOLOGY AND TECTONIC SETI'ING OF THE LATE PERMIAN-EARLY TRIASSIC STEPHENS SUBGROUP, NEW ZEALAND 73

New Zealand. In: A. Panayiotou (Editor), Ophiolites, Proc. Int. Ophiolite Syrup., Cyprus, Rep. Cyprus Geol. Surv. Dep., Nicosia, pp. 480-496.

Dickinson, W.R., 1970. Interpreting detrital modes of gray- wacke and arkose. J. Sediment. Petrol., 40: 685-707.

Dickinson, W.R. and Suczek, C.A., 1979. Plate tectonics and sandstone compositions. Bull. Am. Assoc. Pet. Geol., 63: 2164-2182.

Dickinson, W.R., Beard, L.S., Brakeuridge, G.R., Erjavec, J.L., Fergnson, R.G., Inman, K.F., Knepp, R.P., Lindberg, F.A. and Ryberg, P.J., 1983. Provenance of North American Phanerozoic sandstones in relation to tectonic setting. Geol. Soc. Am. Bull., 94: 222-235.

Enkeboll, R.H., 1981. Petrology and provenance of sands and gravels from the Middle America Trench and trench slope, southwestern Mexico and Guatemala. In: J.S. Watkins, J.C. Moore et al., Initial Reports of the Deep Sea Drilling Project, Vol. 66. U.S. Government Printing Office, Washington, D.C., pp. 521-530.

Forsythe, R.D., Nelson, E.P., Cart, M.T., Keading, M.E., Herve, M., Mpodzis, C., Soffia, J.M. and Harambour, S., 1986. Pliocene near-trench magmatism in southern Chile: a possible manifestation of ridge collision. Geology, 14: 23- 27.

Frost, C.D. and Coombs, D.S., 1987. Nd isotope study of terranes in southern New Zealand. Geol. SOc. N.Z. Misc. Publ., 37A: 47.

Garcia, M.O., 1978. Criteria for the identification of ancient volcanic arcs. Earth-sei. Rev., 14: 147-165.

Grindley, G.W., 1958. The geology of the Eglinton Valley, Southland. N.Z. Geol. Surv. Bull., 58:68 pp.

Harrold, P.J. and Moore, J.C., 1975. Composition of deep sea sands from marginal basins of the northwestern Pacific. In: Initial Reports of the Deep Sea Drilling Project, Vol. 31. U.S. Government Printing Office, Washington, D.C., pp.

507-515. Heller, P.L. and Dickinson, W.R., 1985. Submarine ramp facies

model for delta-fed, sand-rich turbidite systems. Bull. Am. Assoc. Pet. Geol., 69: 960-976.

Howell, D.G., 1980. Mesozoic accretion of exotic terranes along the New Zealand segment of Gondwanaland. Geol- ogy, 8: 487-491.

Hussong, D., Uyeda, S., Blancbet, R., Bleil, U., Elliot, C.H., Francis, T.J.G., Fryer, P., Horai, K., Kling, S., Meijer, A., Nakamura, K., Natland, J.H., Packham, G.H. and Sharas- kin, A., 1982. Site Reports. In: Initial Reports of the Deep Sea Drilling Project, Vol 60. U.S. Government Printing Office, Washington, D.C., pp. 77-411.

Hyden, G., Begg, J.G., Campbell, H.J. and Campbell, J.D., 1982. Permian fossils from the Countess Formation, Mos- sburn, Southland. N.Z.J. Geol. Geophys., 25: 101-108.

Ineson, J.R., 1989. Coarse-grained submarine fan and slope apron deposits in a cretaceous back-arc basin, Antarctica. Sedimentology, 36: 793-819.

Ingersoll, R.V., 1978. Petrofacies and petrologic evolution of the Late Cretaceous fore-arc basin, northern and central California. J. Geol., 86: 335-352.

Ingersoll, R.V., 1979. Evolution of the late Cretaceous forearc basin, northern and central Californi& Geol. SOc. Am.

Bull., 90: 813-826. Ingersoll, R.V., Bullard, T.F., Ford, R.L., Grimm, J.P., Pickle,

J.D. and Sates, S.W., 1984. The effect of grain size on detrital modes: A test of the Gazzi Dickinson point counting method. J. Sediment. Petrol., 54: 103-116.

Johnston, M.R., 1981. Sheet O27AC Dun Mountain (lst ed.). Geological Map of New Zealand 1:50,000. Dep. Sei. Ind. Res., Wellington.

Johnston, M.R. and Stevens, G.R., 1978. New fossil localities in the Maitai Group, Nelson and comments on its age. N.Z. J. Geol. Geophys., 21: 113-115.

Jones, D.L., Howell, D.G., Coney, P.J. and Monger, H.W.H., 1983. Recognition, character and analysis of tectonostratigraphic terranes in North America. J. Geol. Educ., 31: 295-303.

Landis, C.A., 1974. Stratigraphy, lithology, structure and metamorphism of Permian, Jurassic and Tertiary rocks between the Mararoa River and Mt. Snowdon, Western Southland, New Zealand. J. R. Soc. N.Z., 4: 229-251.

Landis, C.A., 1980. Little Ben Sandstone, Maitai Group (Per- mian): nature and extent in the Hollyford-Eglinton region, South Island, New Zealand. N.Z.J . Geol. Geophys., 23: 551-567.

Landis, C.A. and Blake, M.C., Jr., 1987. Tectonostratigraphic terranes of the Croissilles Harbour region, South Island, New Zealand. In: E.C. Leitcb and E. Seheibner (Editors), Terrane accretion and orogenic belts. Am. Geophys. Union, Geodyn. Ser., 19: 179-198.

Le Bas, M.J., 1962. The role of aluminium in igneous clino- pyroxenes with relation to their parentage. Am. J. Sei., 260: 267-288.

Lowe, D.R., 1982. Sediment gravity flows, II. Depositional models with special reference to the deposits of high-density turbidity currents. J. Sediment. Petrol., 52: 279-297.

Lundberg, N., 1982. Evolution of the slope landward of the Middle America Trench, Nicoya Peninsula, Costa Rica. In: J.K. Leggett (Editor), Trench-Forearc Geology: Sedimen- tation and Tectonics on Modern and Ancient Active Plate Margins. Geol. Soc London, Spec. Publ. 10. Blackwell, Oxford. pp. 131-147.

Mackinnon, T.C., 1980. Sedimentologic, petrographic, and tectonic aspects of Torlesse and related rocks. Ph.D. Thesis, University of Otago, Dunedin, 294 pp.

Mackinnon, T.C., 1983. Origin of the Torlessc terrane and coeval rocks, South Island New Zealand. Geol. Soc. Am. Bull., 94: 967-985.

Mutti, E. and Ricci-Lucchi, F., 1972. Le torbiditi dell'Appen- nini settentrionale: introduzione all'analisi di facies. Mere. Soc. Geol. Ital., 11: 161-199.

Nelson, C.H. and Nilsen, T.H., 1984. Modern and Ancient Deep Sea Fan Sedimentation. SOc. Econ. Paleontol. Mineral., Short Course Notes, 14:404 pp.

Nisbet, E.G. and Pearee, J.A., 1977. Clinopyroxene composition in marie lavas from different tectonic settings. Contrib. Mineral. Petrol., 63: 149-160.


Normark, W.R., 1978. Fan valleys, channels and depositional lobes on modern submarine fans: characteristics for the recognition of sandy turbidite environments. Bull. Am. Assoc. Pet. Geol., 62: 912-931.

Pearce, J.A., 1980. Geochemical evidence for the genesis and eruptive settings of lavas from Tethyan ophiolites. In: A. Panayiotou (Editor), Ophiolites, Proc. Int. Ophiolite Symp., Cyprus, Rep. Cyprus Geol. Surv. Dep., Nicosia, pp. 261-

272. Ricci Lucchi, F., 1975. Depositional cycles in two turbidite

formations of northern Appenines (Italy). J. Sediment. Pet-

rol., 45: 3-43. Roser, B.P. and Korsch, R.J., 1988. Provenance signatures of

sandstone-mudstone suites determined using discriminant

function analysis of major element data. Chem. Geol., 67:

119-139. Shanmugam, G., Damuth, J.E. and Moila, R.J., 1985. Is the

turbidite facies association scheme valid for interpreting ancient submarine fan environments? Geology, 13: 234-

237. Stewart, R.J., 1977. Neogene turbidite sedimentation in

Komandorskiy Basin, Western Bering Sea. Bull. Am. As-

soc. Pet. Geol., 61: 192-206. Stewart, R.J., 1978. Neogene volcaniclastic sediments from

Atka Basin, Aleutian Ridge. Bull. Am. Assoc. Pet. Geol.,

62: 87-97. Turnbull, I.M., 1979a. Stratigraphy and sedimentology of the

Caples terrane of the Thompson Mountains, northern

Southland, New Zealand. N.Z.J. Geol. Geophys., 22: 555- 574.

Turnbull, I.M., 1979b. Petrography of the Caples terrane of the Thompson Mountains, northern Southland, New Zealand. N.Z.J. Geol. Geophys., 22: 709-728.

Turnbull, I.M., 1980. Structure and interpretation of the Ca- pies terrane of the Thompson mountains, northern South-

land, New Zealand. N.Z.J. Geol. Geophys., 23: 43-62. Van der Plas, L. and Tobi, A.C., 1965. A chart for the

reliability of point counting results. Am. J. Sci., 263: 87-90. Walker, R.G., 1975. Generalised facies models for resedi-

merited conglomerates of turbidite association. Geol. Soc. Am. Bull., 86: 737-748.

Walker, R.G., 1967. Turbidite sedimentary structures and their relationship to proximal and distal environments. J. Sedi- ment. Petrol., 37: 25-43.

Walker, R.G. and Mutti, E., 1973. Turbidite facies and facies associations In: G.V. Middleton and A.H. Bouma (Editors), Turbidites and Deep Water Sedimentation. Pac. Sect., Soc. Econ. Paleontol. Mineral., Short Course Notes, pp. 119-157.

Waterhouse, J.B., 1964. Permian stratigraphy and faunas of New Zealand. N.Z. Geol. Surv. Bull., 72:101 pp.

Waterhouse, J.B., 1979. A new species of Permophorus Chavan (Bivalvia) from the Triassic of New Zealand. N.Z.J. Geol. Geophys., 19: 373-385.

Yerino, L.N. and Maynard, J.B., 1984. Petrography of modem marine sands from the Peru-Chile Trench and adjacent areas. Sedimentology, 31: 83-89.

Sedimentology and tectonic setting of the Late Permian-early Triassic Stephens Subgroup, Southland, New Zealand: an island arc-derived mass flow apron

Documents