TECTONICS, VOL. 13, NO. 6, PAGES 1327-1344, DECEMBER 1994
Rates of deformation, uplift, and landscape development
associatedwith active folding in the Waipara area of North
Canterbury, New Zealand
Andrew Nicol •
Departmentof Geology,Universityof Canterbury,Christchurch,
New Zealand
Brent Alloway
Department
of Geology,AucklandUniversity,TamakiCampus,Auckland,New Zealand
Philip Tonkin
Departmentof Soil Science,LincolnUniversity,Canterbury,New Zealand
Abstract. Analysis of the geometryand ages of faulted and
tilted late Quaternary fluvial terraces and their associated
coverbedsprovideevidenceof active folding at three localities in the Waipara area of North Canterbury,New Zealand.
Terracesurveydata,the occurrence
of the approximately22.6kyr-oldAokautereAsh, andexaminationof soil profiles indicatethat folding hascontinuedinto the late Holocenebut that
the amountsand rates of deformation are locally variable.
Ratesof uplift in the Waiparaareaarecomparedwith thosederived from marine terracespreservedat the Pacific coast,east
of the studyarea. Resultsindicatethat ratesof measurabledeformationreach a maximum along the Waipara range front,
where bedrock deformation is most intense and shortening
last 0.8 + 0.4 m.y. since the onsetof Quaternarydeformation
in the Waipara region.
Introduction
The fundamentalprerequisitefor documentingactive fold
growthis the recognitionof late Quaternarymarkerhorizons,
whichareof sufficientageandlateralextent to display gradientsthatdepartfrom their initial orientation.In New Zealand
many suchmarkershave been used to documentfolding;
theseincludemarineterraces[e.g.,Singh, 1971; Ghani, 1978;
Berryman,1993a, b], fluvio-glacialterraces[Suggate,1987],
andwave-planersurfaces[Lewis,1971]. Althoughthe analyratesof up to 5.57 + 0.69%/100 kyr occur.Acrossthe coastal sis of active folding is often difficult owing to the continuous nature of deformation and the absence of suitable markrangesthe averagerate of shorteningis 0.8 + 0.4%/100 kyr,
ers,it mustbe recognizedthat folding may accountfor a sigwhich correspondswith an absoluteshorteningrate of 1.4 +
nificant
componentof the total strain. In this context the
0.6 m/kyr and representsonly a small proportionof the predictedplate motion vector in this region. Uplift rates range analysisof active folding is extremelyimportantif rates of
from 0-1.83 m/kyr for a late last glacial fluvial terrace and deformationandpotentialseismichazardareto be adequately
alsosuggestthat fundamental
from 1.36-2.16 m/kyr for three marine terraces.Fluvial and characterized.Our observations
marineterraceuplift ratesvary in accordwith the geometries relationshipsexist betweenlandscapedevelopmentand the
distribution of folds in areas of active deformation.
of the folds in bedrock,and the spatialpattern of uplift diIn an attemptto characterizeactive folding in the Waipara
rectly reflectsfold growth. The structurecontourpattern of
area
of North Canterbury,New Zealand(Figure 1) we have infolded surfacesprovidesa first approximationto the spatial
tegrated
the analysisof bedrockstructure,Quaternarystratigpatternof uplift. Differential uplift due to folding accounts
and pedology.Data have been colfor up to approximately55-75% of the total uplift and has raphy,geomo•10hology,
lected from three different localities across the Waipara
producedfolds with structuralrelief of about 1300 m (i.e.,
amplitudesof 600-700 m). Thesefolds have formed over the Syncline,nearthe townshipof Waipara(Figure 1). Fold geometrydata,principallyderivedfrom altitudinal surveying,
include disruptedcover bed stratigraphyand deformedfluvial terracesurfaces.Geometricdataare complementedby age
I Nowat EarthSciences
Department,
University
of Liverpool,determinations
providedby theradiometrically
datedand laterally extensiveAokautereAsh, a radiocarbondate and detailed analysisof soil profile morphologies.The studyhas al-
Liverpool,England.
lowed conclusionsto be drawn about the rates of shortening
anduplift associated
with folding,the ageof coverbeds,flu-
Copyright1994by theAmericanGeophysical
Union.
vial terrace surfaces and their deformation, and the relation-
Papernumber94TC01502.
ships of active folding to landscapedevelopmentin the
0278-7407/94/94TC-01502510.00
coastalrangesof North Canterbury.
1327
1328
NICOL
I
•
ET AL.'
ACTIVE
FOLDING
IN WAIPARA
AREA OF NORTH
CANTERBURY
I
•
•o
•
"•
o
;
u
I
I
I
ß
.
.I
, .'
'
i
/
ß!
!
I
I
ß
ß
i
ß
!:,
! ßi
ß
NICOL
ET AL.:
ACTIVE
FOLDING
IN WAIPARA
AREA
OF NORTH
CANTERBURY
1329
peatedlymigratedacrossa floodplaindominatedby a mosaic
of grassland-shrubland.
These sedimentswere inferred by
Structure
Harristo correlatewith those depositedduring and since the
last glacial maxima (approximately 12 - 24 kyr ago; [see
North Canterburyis located on the southeasternmargin of
Pillans et al., 1993] basedon the apparentabsenceof loess,
the Marlborough Fault System (MFS), a portion of the
obliquelyconvergentNew Zealandplate boundary dominated glacial, or periglacial deposits,the slight degreeof weathering, and lack of stream dissection.This estimate was supby right-lateral strike-slip faulting. The style of faulting at
ported by a conventionalradiocarbonage of 10,550 + 150
the surfacein North Canterburyis dominatedby reversefaults
and thruststhat mainly strike northeastand dip to the south- yearsB.P. obtainedfrom a Hyridella shell sample within an
uppersilt unit, but suchshell samplescan be subjectto coneast. These faults are closely associatedwith a number of
tamination,and we regard this date as a minimum. Although
asymmetricmacroscopicfolds, whichare developedwithin an
approximately1-kin thick sequenceof late Cretaceous-early the uppermostportion of the meanderstream sequencewas
thoughtto representdepositionduring the last glacial maxPleistocenesedimentaryrocks. The folds are frequently charthe entiremeanderstreamsedimentarysuccession
acterized
bysteep
northwest
limbs
(600
øtooverturned
dips) ima (LGM),progressive
infilling of a basinalong the hinge of
and shallow southeastlimbs (10-30 dips), with the steep represents
the WaiparaSynclineduringthe late Quatemary.For the purlimb often being faulted. Collectively, the faulting and folding have accommodatedapproximately 12-15% northwest- posesof this studythe late last glacial aggradationsurfacein
the Omihi-Waiparavalley will subsequently
be referredto as
southeastregional shorteningbetween the Pacific coast and
the "Omihi Surface"(seeFigure 1 for lateral extent).
Hope Fault [Nicol, 1991], the southernmostelement of the
The Oraibi aggradationsurfaceis underlain by up to apMFS.
proximately
3 m of calcareous
silts,sands,and gravels,which
The topography is dominated by a series of northeast
may containor overliea silicic tephra(Figure2) eruptedfrom
trending valleys and ridges that reflect the influence of the
underlyingstructures.The ridges are composedof indurated the TaupoVolcanic Zone, which is here correlatedto the apMesozoic graywacke and argillite
basement (Torlesse proximately22.6-kyr-old AokautereAsh (see next section).
Supergroup)and representthe exposedcoresof macroscopic The presenceof thistephrabelow the Oraibi Surfacesuggests
that the terrace is necessarilyyounger than approximately
anticlines,which are generally confined to the hanging walls
22.6 kyr. The maturenatureof the soil profiles(vertic melanic
of the faults. The valleys are locatedon the footwalls of the
soils,seeFigure5) suggestthat the surfaceis unlikely to be
faults, coincident with the major synclinal structures.The
Holocene
in age.We infer that the terracesurfaceformeddurlargest of these valleys is the Culverden Basin, a structural
ing
late
last
glacial (approximately12 + 2 kyr ago) aggradadepressionthat, like many similar structures in North
tion of Omihi Stream.The Canterburyand Omihi aggradation
Canterbury,containsan unknown,but probably small (<200
surfacesappearto coalesceand gradeto the sameheightin the
m) thicknessof Quaternary gravel deposits.Active faults
generalvicinity of Waiparatownship(Figure 1).
boundthe northwestside of the ridges and displaceboth geThe fourth geomorphicsurfaceis Weka Fan, characterized
omorphic surfacesand Quaternary gravels [Wilson, 1963;
by a series of narrow alluvial ridges that extend east and
Gregg, 1964;Maxwell, 1964; Cowan, 1992a].
southeastfrom the mouth of Weka PassacrossState Highway
Structuresin the Waipara area (Figure 1) constitutepart of
1 (Figure 1). Southeastof Weka Pass,the Weka Fan is clearly
the North Canterburyfold and thrust (or reverse fault) belt.
higher
than the older Canterbury Surface, and lobes of alluLocal folds are characterizedby two anticline-synclinepairs
that strikeapproximatelyNNE and vergewestward.The most vium overlie loess which was presumablyderived from the
CanterburyGravels.A possiblecorrelative of this loess-mandominantof these structuresare the Waipara Syncline and
tied surfaceoccursas a small, eastwarddipping remnantpreBlack Anticline, which are separatedto the northby the Omihi
North Canterbury Geology
Fault, a major range-bounding structure. Yousif [ 1987]
showedthat folds immediately adjacentto the study area, includingthe Black Anticline, are currently active. Active folding extendswestwardacrossthe Waipara Syncline for at least
15 km [Campbelland Nicol, 1992].
Geomorphic Surfaces
In the Waiparaarea,five groupsof geomorphicsurfacesare
identified using soil morphology, elevation, and available
age data. The oldest, named the Teviotdale Surface [Wilson,
1963], is of small areal extent and characterizedby isolated
loess-mantledremnantspreservedto the south and southeast
of the WaiparaRiver. This surfaceis not studiedhere.
The next youngestand more extensivesurfacewas named
"CanterburySurface" [Wilson, 1955], and the gravels that
immediately underlie this surface were referred to as
"Canterbury Gravels". Harris [1982] recognized that
CanterburyGravelswithin the Waipara-Omihivalley were depositedby a slow flowing, meanderingriver channelthat re-
servedon the easternside of the Mound (Figure 1).
Below the Canterburyand Omihi Surfaces,flights of river
terracesstepdown to the active floodplains.These surfaces
are characteristicallystony, with rudimentary soil profiles
that are predominantly<0.25 m in thickness(orthic brown
soils,Figure 5) and are interpretedas degradationterraces.A
radiocarbondate from charcoal sampled from beneath a
degradationterracein LimestoneCreek (Figure 1) produceda
conventional radiocarbon age of 1008 + 63 years B.P.
(NZ7957). The sample is inferred to date ponding of
Limestone Creek behind the secondhighest Waipara River
terrace.This, togetherwith the immature soil profiles, suggeststhat degradationof the Waipara River from the second
highestterraceto its presentlevel (approximately20 m), occurredduringthe late Holocene.
Occurrence and Recognition of the Aokautere Ash
The largestandmostwidespreadlate Pleistocenesilicic air
fall tephra,AokautereAsh Member of the KawakawaTephra,
1330
NICOL ET AL.: ACTIVE FOLDING IN WAIPARA AREA OF NORTH CANTERBURY
Site B
Site A
Omihi
Limestone
Stream
Creek
(N34/924935)
(N34/915943)
WekaFanalluvialgravels
(dominantly
Torlesse
lithologies)
well sorted
medium
sand
laminated to weak X-bedded
(mm-dcm)
well sorted, calcareousfluvial
silts and sands.
•c
c c cl
'_••
c.5cm
thick
Fluvial gravels
containingmixed
•x x •2x• • •,
SILICIC
TEPHRA
BED
I•;,d•
(distal
correlative
ofAokautere
Ash
•x..•
)
•
["•'.•
•
Amuri
laminated
toweak
•_
•
Limestone
andTorlesse
lithologies
X-bedded
(mm-dcm)
wellsorted,
calcareous
fluvial
silts
andsands
with
[...:?..•:_:'.•;..':• moderately
sorted
gravel
interbedscontainingmixed
Amuri Limestone and Torlesse
SILICIC TEPHRA
lithologies
BED
weaklylaminatedto
massive fluvial silts and
fine sands
KEY
Siltloam
Sandy
loam
Fluvial gravels
containingmixed
Loamy
sand
Amuri
.•-•
•-•
Sandy
loamclay
Clay
loam
Limestone
andTorlesse
lithologies
10
Clay
!ccc]
Free calcium
carbonate
Stones
and/
or gravels
KOWAI
FORMATION
Figure2. Stratigraphy
enclosing
theAokautere
Ashcorrelative
at siteA alongOmihiStreamand siteB along
LimestoneCreek,Waipara(seeFigure1).
NICOL
ET AL.:
ACTIVE
FOLD1NG
1N WAIPARA
was eruptedfrom the Lake Taupo area (see Figure 1 inset) at
approximately22.6 kyr ago [Wilson et al., 1988; Froggatt
and Lowe, 1990; Pillans et al., 1993]. This air fall tephra has
proven to be extremely useful in regional studies becauseit
defines an isochronoushorizon for correlation and age control of LGM loess,peat,and colluvial deposits,as well as geomorphic surfacesand associatedsoils. This air fall tephra is
found extensively within central and southernNorth Island
[Milne, 1973; Palmer, 1982], as well as northern South Island
[Eden, 1983; Campbell, 1986]. Isolated occurrenceshave also
been reported from western South Island [Mew et al., 1986];
however, in North Canterbury,AokautereAsh has only been
previouslyreportedfrom one site just southeastof the study
area, near the townshipof Amberley [Kohn, 1979]. Here the
tephrawas fission track dated at 20,300 + 7100 years B.P. In
other areas of central and southern South Island, where this
tephra is absent,age determinationsof late Quaternary geomorphic surfacesand/or soils have been mostly restricted to
AREA
OF NORTH
CANTERBURY
1331
AokautereAsh is the only macroscopic silicic tephra extensivelypresentin late Quaternarystrata. Consequently,its
thickness and stratigraphic position within LGM loess in
most areasis usually distinctive enoughto permit its reliable
identification.However, in the Waipara area the silicic tephra
occursenclosedwithin late Quaternaryfluviatile deposits but
has yet to be found as a discretelayer within LGM loessof the
area; its identity as Aokautere Ash therefore required confirmation by glassshardchemistry.
Glass shardsof samplesfrom two Waipara localities were
analyzed for major elementsusing electronmicroprobe(EMP)
analysis.The EMP analyses(Table 1) confirm that this silicic
tephra cannot be differentiated from similar analyses of
AokautereAsh obtainedfrom its type sectionand a number of
distal localities. However, these analysesare clearly distinguishablefrom other widespread Pleistocenesilicic tephras
erupted from the Taupo Volcanic Zone, as illustrated in the
CaO-FeO-K•O
ternarydiagram(Figure3).
usingt4Candthicknesses
of weathering
rindsongraywacke
sandstone clasts [e.g., Chinn, 1981; Knuepfer, 1988;
Campbell and Nicol, 1992].
In this study, four new localitiesof a silicic tephra are identified and described in the vicinity of Waipara in North
Canterbin3
r (Figure 1). Here this tephra occurstypically as an
intermittently exposedapproximately<0.05-m thick horizon
encapsulatedwithin fine-grained, calcareous fluviatile deposits (Figure 2). This tephra can be readily distinguished
fromsimilarlythick,very palebrownto whiteCaCO3 layers
on the basisof its characteristicpinkish color (SYR 7/3) and
silt-grade,glassytexture.
Evidence for Active Folding
Fault Bank Locality
This locality was first described by Harris [1982]. Here,
approximately10-20 m of Quaternaryfluviatile bedsrest with
angular unconformityon Plio-PleistoceneKowai Formation
bedrock(Figure 4; see Figure 1 for location). These fluvial
beds, characterizedby pebbly to clayey sedimentsderived
principally from post-Torlesselithologies, are both faulted
Table 1. AverageMajor ElementComposition
of GlassShardsFrom AokautereAsh at Its Type Section
ComparedWith Its Distal Correlatives
Aokautere
AshMember
oftheKawakawa
Tephra
Formation
TypeSection•
SiO2
A1203
TiO2
79.01 ñ 0.38
12.15 ñ 0.30
0.13 ñ 0.03
Taranaki a
79.35 ñ 0.30
12.40 + 0.12
0.12 ñ 0.03
Amberleya
78.81 + 0.44
12.44 ñ 0.15
0.11 ñ 0.03
Waipara
Waipara
ThisStudy
b,
ThisStudy
b,
Site A
Fault Bank
77.55 ñ 0.43
12.65 ñ 0.26
0.15 ñ 0.05
77.73 + 0.36
12.48 ñ 0.23
0.17 ñ 0.05
FeO
1.15 + 0.10
1.19 + 0.11
1.18 ñ 0.08
1.20 ñ 0.13
1.16 + 0.12
MgO
0.12 + 0.02
0.12 ñ 0.01
0.12 ñ 0.02
0.15 ñ 0.03
0.20 ñ 0.07
CaO
1.05 + 0.09
1.07 ñ 0.09
1.02 ñ 0.09
1.08 ñ 0.09
1.10 + 0.09
Na20
K20
3.31 + 0.21
3.07 + 0.25
2.47 + 0.37
3.08 + 0.10
3.35 + 0.22
2.96 + 0.37
3.87 ñ 0.19
3.12 + 0.17
3.87 + 0.26
3.07 + 0.15
0.23 + 0.03
0.22 + 0.06
6.10 + 1.04
5.48 + 1.26
C1
--
H20
n
5.65 + 1.62
0.18 + 0.02
7.14 + 0.98
74
15
--
3.41 + 1.40
11
19
21
All majorelementdeterminations
weremadeon a JEOL JXA-733 electronmicroprobehousedat
Victoria Universityof Wellington.Analyseswere determinedusinga beamcurrentof 8 nA, beamdiameterof 20 mm, and3 x 10-speak counts(meaned).Valuesare in weight-percentoxide,recalculatedto
100% on a fluid-freebasis.H20 is by differencefrom 100%;total iron expressed
asFeO; n is numberof
analyses.The majorelementcomposition
of internalglassstandards
KN-18, VG-99, andVG-568 were
routinelyanalyzedto checkandcorrectfor machinedrift. SeeFigure2 for SiteA andFigure3 for Fault
Bank.
a AnalystwasP. Froggatt.
b Analyst
wasD. Manning.
1332
NICOL
ET AL.:
ACTIVE
FOLDING
IN WAIPARA
AREA
OF NORTH
CANTERBURY
FeO
v
v
v
v
CaO
K20
ß
Aokautere Ash distal correlative
+
(fromAtlowayetaL, 1994)
Aokautere Ash distal correlative
Waipara-- Fault Bank
AokautereAsh from typesection-WhangamataRoad,CentralNorth Island
•
distalAokautereAsh-
ß
Rangitawa
Tephra(350ka)typesection,southernNorth Island
x
PotakaTephra(1.05Ma) -typesection,southernNorth Island
(fromAtlowayetal., 1993)
[]
Taranaki, western North Island
(fromAltowayet al., 1994)
A
distalRotoehuTephra(55 ka) -Taranaki, western North Island
Waipara -- Site A
O
v
(fromAtlowayetal., 1993)
distal Aokautere Ash --
Amberley,east-central
SouthIsland
Figure3. FeO-K20-CaO
ternary
plotshowing
thecomposition
ofAokautere
Ashcompared
withseveral
other
widespread
Pleistocene
tephraeruptedfromthe TaupoVolcanicZone. Ticks on the diagramare at 10% intervals.
and tilted with intense local deformation adjacent to a
mesoscalefault. Away from the fault the fluvial beds thicken
and contain the approximately 22.6-kyr-old Aokautere Ash.
Structurecontoursconstructedon the top of the 0.05-m thick
ashhorizon
indicate
thatit dipsatapproximately
9ø(Table2)
SSW but has a much reduced apparentdip in the bank exposureof Figure 4. Below the ash horizon the beds becomeincreasinglytilted, fi'om which we infer that the cover beds increasein age toward the unconformitysurfaceand that the onset of folding began well before cover bed deposition. To
qualify theseage relationships,we have usedratesof shortening for the period sinceash depositionand dip data from the
pre-ashbeds(seeTable 2 for furtherdetails).
The fault exposedat the Fault Bank locality (Figure 4) is a
mesoscalestructurethat strikessoutheastat an oblique angle
to the main folds. Slickenside striationson the fault plane
suggestthat it has accommodateddip-slip movement, while
two gravel marker horizons,G1 and G2, have been displaced
in a reversesenseby 3.8 and 5.7 m, respectively (Figure 4).
The fault has accommodateda total of at least 5.7 m displacement during two or more eventssincethe deposition of the
top of the G2 gravel bed. The top of the G2 gravel unit is
eroded,so this figure is a minimum. Aokautere Ash perceptibly thins toward, and is apparently deformed adjacent to the
fault, which suggeststhat the last fault movement postdates
approximately22.6 kyr. Using the age derivedfor the top of
the basal gravel bed (approximately 78 + 19 kyr, see Age
Estimatessection), a minimum slip rate of 0.073 + 0.019
m/kyr is calculatedfor the fault.
Shortening Calculations. The age and dip of the Aokautere
Ash are used to calculate both the amount and rate of local fi-
nite shortening for the period since ash deposition at the
Fault Bank locality. These calculationsare basedon the following assumptions:(1) the ash horizon and all beds for
which ageshave been calculatedwere approximatelyhorizontal prior to deformationand (2) the ash horizon and bed dips
are indicative of folding achieved by bed rotation about a
horizontal axis associatedwith constantrates of shortening.
The first assumptionappearsto be reasonable,given that
the ash horizon is containedwithin fluvial pebbly silts and
sands,
where
present
floodplain
gradients
øforOmihiStream
and the Waipara River are generally <0.3 . The mechanicsof
fold developmentare not addressedin detail here. However,
given that the folds are developed in late Cretaceousand
younger strata, which generally do not exceed 1.5 km in
thickness,and that the active folding we are describingis now
occurring at the ground surface, we conclude that the folds
formed at shallow levels (<1.5 km) in responseto buckling.
The absenceof abundantslickensidesand/or striated bedding
surfacessuggeststhat at an outcrop scale, bedding plane slip
NICOL ET AL.' ACTIVE FOLDING IN WAIPARA AREA OF NORTH CANTERBURY
! I'!
ß
o•
1333
1334
NICOL
ET AL.: ACTIVE
FOLDING
IN WAIPARA
AREA
OF NORTH
CANTERBURY
Table 2. Age Estimatesfor Cover Beds, the Bedr•x:kStrathSurfaces,and the Onsetof
Deformationat the Fault Bank Locality
,
Dated Feature
Dip, deg,a
Shortening, %
Age, kyr
9.1
(SCD)
1.26 + 10.3
22.6 + 0.3
1.26 + 11 to
<22.6 + 0.3 to
2.2 + 16.7
39.3 + 11.4
4.4 + 11.8 to
78.5 + 18.9 to
5.7+ 10.3
103.1 + 23.3
5.7 + 10.3
103.1 + 23.3
34.7+ 4.1
622.1+ 101.9
d
1, AokautereAsh
2, Fluvial sandsand silt unit
(envelopes
Aokautere
Ash)b
3, Gravel unit
(abovebedrock
strath)
4, Bedrock strath
<9.1
to 12.0
(SCD)
(4c)
17.0 to 19.5
(2c)
(4c)
19.5
(4•)
5, KowaiFormation
49.2
(5c)
The agesare derivedusingthe orientationandageof the AokautereAsh to determinethe rate
of shortening(5.57 + 0.69%/100 kyr) sinceashdeposition(approximately22.6 + 0.23 kyr). The
total shortening,calculatedfrom the bed dips, is dividedby the rate of shorteningto estimatethe
age of eachbeddedunit. Uncertaintiesin bed dip measurements
andthe age of the ashare usedto
derivethe errors.The locationof eachnumberedunit is markedon Figure 4 (circlednumbers).
a Errorsare + 2ø. SCD is structurecontourderived,with error of + 1ø.
bThisunitwaspreserved
onlyontheSWsideofthefault(seeFigure4).
c Valuesrepresentnumberof dip measurements
usedto calculatethe averagedip values.
dValuesareinferred
topdatetheonsetof deformation,
notdeposition
of formation.
is not significantin the study area, except adjacentto faults
and within mudstonelithologies that dip at angles >60 . The
folding is thereforemodeledby assumingthat departureof
bedding dips from the horizontal representsa bed rotation
which hastaken place abouta confined horizontal fold hinge
zone, with little or no changein bed length. In the Waipara
o
area,
foldhinges
arecommonly
subhorizontal
(<15ø)andthe
useof horizontalaxesof bed rotation appearsto be a reasonable approximation. In order to provide age estimates,we
have also assumedthat folding reflects constantshortening.
This assumptionof constant shortening representsa departure from previous studies [e.g., Wellman, 1971; Ghani,
1978], where constant rates of tilting have been inferred.
However, there is almost no mechanism for accommodating
constanttilting during fold growth, while it is at least possible that folding reflectsconstantshortening.Thereforefor the
purposesof this study, invariant shortening is assumed,although over the entire period of late Cenozoic deformation,
this is likely to representan oversimplification.
Shortening is calculated for a given bed length using the
equation
S = 100(1- cos 0),
where$ = shortening(in percent)and 0 = beddingdip. Using
the AokautereAsh horizon data (Table 2, top row), both the
amount and rate of incrementalshorteningare calculated for
the period sinceashdeposition(approximately22.6 kyr B.P.).
The rate of shortening (5.57 ñ 0.69%/100 kyr) is subse-
quentlyextrapolatedback in time to derive age estimatesfor
the variably dipping fluvial beds below the ash (Table 2).
This extrapolationalso providesa first-orderestimatefor the
ageof the onsetof deformation,which is observedin the underlying Kowai Formation bedrock (Table 2). Our age estimates for the fluvial depositsare supportedby independent
data, while the inferred age for the onset of deformation
shouldbe regardedas a crude estimate.These data are presented in Table 2.
Age Estimates. Assuming constantshortening,all of the
deformation
observed
within
the fluvial
beds can be accom-
modated over a period of approximately 100 kyr. On the
downthrown and northeast side of the fault (Figure 4), only
basal parts of the fluvial stratigraphyare preservedand the
Aokautere Ash is absent. To the southwest of the fault, local-
ized hanging wall uplift has condensedthe sequence,and
from the markedreductionin beddingdips acrossthe erosion
surface(Figure4) we infer that most of the 39- to 78-kyr-old
stratigraphyis not preserved.It is suggestedthat the fluvial
beds between the ash horizon
and the erosion surface below
rangein age from approximately22.6 to 39 kyr, while the remainder of the beds date between approximately78 and approximately 103 kyr (Table 2). The age of the intervening,
faulted, and unconformity-boundedgravel unit (G1, Figure 4)
is somewherein the range of 39 to 78 kyr and can be correlated acrossthe fault where a gypsum-bearingcarbonaceous
horizon is found to directly overlie the gravel unit (G1). On
the basisof the identificationof a cold (full glacial) climate
pollen spectrum (M.S. McGlone, personal communication,
NICOL
ET AL.:
ACTIVE
FOLDING
IN WAIPARA
AREA
OF NORTH
CANTERBURY
1335
1992) sampledfrom the carbonaceous
horizon (Figure 4), and from 0 to 18.5 m. Assuming an estimatedage of approxiage constraintsalready calculatedfrom bracketing the sedi- mately 12 + 2 kyr for the Omihi Surface, uplift rates ranging
from 0 to 1.54 m/kyr (-+-approximately20%) are calculatedfor
ments on the southwest side of the fault it is suggestedthat
the carbonaceoushorizon and underlyinggravel unit accumu- six pointson the terrace(Figure7). However,if the gradientof
lated during an episode of cold (full glacial) climate that the aggradationsurfaceis extrapolatedto the hinge of the adbroadly correspondswith oxygen isotope 4 (approximately jacent anticline,then the calculateduplift rate would be 1.83
+ 0.52 m/kyr. This figure is remarkablycloseto the uplift rate
70 kyr ago).
(1.70 m/kyr)derived by Yousif[1987] for the same anticline
A marked angular unconformityat the base of the fluvial
hinge 0.5 km to the south in Yellow Rose Creek (Figure 7).
beds on the southwest side of the fault signifies that the
Kowai Formationbedrockis appreciablymore deformedthan Tilting of the Omihi Surface reflects shortening rates of
0.15%/100 kyr, considerably less than that calculated from
the overlying Quaternarybeds. Accordingly, the age for the
onsetof deformationof bedrockis inferredto be considerably the Fault Bank locality where the influenceof the Omihi Fault
older than the calculated age of the bedrock strath is likely to be greater, but similar to the rates of shortening
(approximately103 + 23 kyr). Using the rate of shortening that couldbe inferredfrom bed dips of 5 - 10ø formed in the
last 500-700 kyr (see Timing and Strain Rates of Deformation
derived from the Aokautere Ash, all of the observable deforsectionfor further discussionof strainrates).
mation of the Kowai Formation at this locality could have
In contrastto the Omihi Surface,the Waipara River degrabeen achieved in the last approximately500 to 700 kyr (for
dation terrace profiles display considerably less departure
further discussion, see Timing and Strain Rates of
from the presentfloodplain gradient(Figure 5). This suggests
Deformation section).
that if the rate of deformation has remained approximately
constant, then the degradation terraces are considerably
Glenray Locality
younger than the aggradationterrace. The radiocarbondate
A flight of fluvial terraceson the north bank of the Waipara
(1008 + 63 yearsB.P.), which is inferredto date the ponding
River are variously tilted owing to folding associatedwith
of LimestoneCreek duringthe formation of the secondhighthe developmentof the Waipara Syncline and adjacent antiest Waipara River degradationterrace, implies that those tercline structures(Figure 1). The terracesurfacesdisplay an inrace surfaces lower in elevation than the dated terrace are less
creasingdeparturefrom the presentstreamgradientswith verthan 1000 years old. A secondradiometricdate derived from
tical distanceabovethe Waipara River floodplain. The terrace
wood also sampledfrom within "ponded"deposits,but in a
gradientsare measuredfrom profiles surveyedby automatic
streamimmediately adjacentto LimestoneCreek, produceda
level, normalto the strike of folding (Figure 5). Sevenprinsimilar age of 1150 + 55 years B.P. (N34/f76 [Yousif, 1987])
cipal terracesare recognizedabove the presentfloodplain of
and supportsthe late Holocene age assignedto the terraces.
the Waipara River, five of which are displayed in Figure 5.
Theseradiocarbondatesare further supportedby weatheringTheseterracesare divided into two groups (1) the uppermost
rinddataand•4CdatesderivedfromWaiparaRiverterraces
Omihi aggradationsurfaceand (2) a flight of Waipara River
upstream[Campbell and Nicol, 1992] and require that the
degradationsurfaces.The measurementerror associatedwith
river at Waiparahas downcut approximately20 m during the
surveying is conservatively estimatedat + 0.1 m, while local
late Holocene.
topographyon the terrace surfacesis of the order + 0.5 m.
Consequently,consistentdeparturesof terrace surfaceelevations in excessof 0.6 m, which can be correlated with the locations and geometry of the underlying macroscopicfolds, are
inferred to reflect differential uplift associatedwith folding.
The most striking aspectof the terraceprofile data (Figure
5) is the marked eastwardclimb of the Omihi Surfacetoward
the hinge of the adjacentanticline. To the west the surfacehas
an elevationof 70.6 m abovemeansea level (amsl) and is approximatelyparallel to boththe degradationsurfacesand the
Waipara River floodplain. Between 1.4 and 1.6 km from the
anticline hinge the aggradationsurface begins to increasein
elevation,reachinga maximum elevation of 85.2 m (amsl), at
a distanceof 0.3 km from the hinge.This local differential uplift and the resulting tilt on the Omihi Surface appearsto directly reflectgrowth and developmentof the anticline immediately west of the surface(Figure 1). By assumingthat the
Omihi Surfacewas initially horizontalin the plane of the profile, a reasonableassumptiongiven that the surfaceprobably
developedowing to southwardaggradationand the profile is
orientedeast-west,then it is possibleto calculate the amount
of local uplift at any point along the surface.The amountsof
uplift are calculatedrelative to the Waipara Syncline hinge
(the lowestpoint in the structure),which is assumedto represent the point of zero local (or relative) uplift and to range
Mound Locality
The Mound locality is a loess-mantled remnant of
TeviotdaleSurface[Wilson, 1963] which rises approximately
25 m above the younger and laterally extensive Canterbury
Surfaceof late last glacial age (Figure 6a). The flanks and top
of the Mound are characterizedby remnantsurfaces,which are
tilted to the east.Below the CanterburySurface,Waipara River
degradationhasresultedin the developmentof a seriesof terrace surfaces,which step down to a large (approximately 20
m) vertical river bank scarpabovethe present-day floodplain.
Exposedin this scarp,immediatelybeneaththe Mound, is a
sequenceof gravel units which displays increaseddeformation with age.Terraceheights and a sectiondiagram are plotted in Figure 6a, which is constructedalmostperpendicularto
the strike of local folding.
Figure6a clearly illustratesthe presenceof an asymmetric
anticline, with westwardvergence, developed within the se-
quenceof gravelsof presumed
lateQuaternary
age.This structure was identified by Wilson [1963] and appearsto be the
northern continuationof a fold with a similar geometry imaged severalkilometers to the south in seismic line 1A of
Kirkaldy and Thomas[1963]. The seismicallyimagedfold is
1336
NICOL ET AL.' ACTIVE FOLDING IN WAIPARA AREA OF NORTH CANTERBURY
%
i
I
I
I
I
I
i
I
(Ismem) UoT.
le,Xai3
NICOL ET AL.' ACTIVE
FOLDING
IN WAIPARA
AREA OF NORTH CANTERBURY
o
-8
-8
o
1337
1338
NICOL ET AL.' ACTIVE FOLDING IN WAIPARA AREA OF NORTH CANTERBURY
closely associatedwith a reverse fault that displaces the
steepestlimb of the asymmetricfold. We therefore infer that
folding at the Mound Sectionreflects the presenceof a macroscopicreversefault at depth (crosssection,Figure 1).
The post-Canterburyterrace surface(T3) with its thin veneer of degradationgravels appearsundeformed(Figure 6a).
Its strath at the anticlinal hinge unconformably overlies
severelyshearedpre-Canterburyweatheredgravels;whereas,
immediatelyeither side of the anticline, the post-Canterbury
gravel strath overlies Canterbury gravels paraconformably.
Similarly, the Canterburyterracesurface(T2) acrossthe anticlinal structurealso appearsundeformed,but aggradational
gravelsbeneaththis surfacethin perceptibly acrossthe hinge.
Beneath the Canterbury gravels, two older gravel units are
recognized and display variable dips in the vicinity of the anticlinal hinge but become increasingly parallel short dis-
tances
oneithersideof thehinge.Intense
foldingandshort-
ening
areconfined
toanarrow
zone
which
extends
upto 200-
300
moneither
side
ofthe
anticline
hinge.
The disparityin dip betweenundeformed
and deformed
gravel
unitsattheMound
Section
could
result
if therateof
deformationhas remained relatively constantsince deposition and the age of eachdeformedgravel unit is markedly different from the age of gravelsassociatedwith undeformedterrace surfaces.The inference of late Quaternary deformation is
supported by the geometry and inferred age of a preCanterburyterrace surface (T1, Figure 6a), located immediately east of the Mound, where its elevation relative to the
underlying T2 terrace declines to the east until the two surfaces converge. East of this point, the T1 terrace is not preservedand appearsto have been buried by the youngerT2 and
T3 terracedeposits(Figure 1). Tilting of T1 terrace is consistent with that of the underlying pre-Canterbury gravels, but
T1 is coveredby at least one loess sheet (approximately 3 m
thick),
thepresence
ofwhich
implies
anageofatleast
20-30
kyrold.This
loess
sheet
canbetraced
east
from
theanticline
where
itdips
beneath
the
Canterbury
gravel
strath.
Wetherefore concludethatthe sequenceof variouslydeformedgravel
unitsareindeedsignificantly
olderthantheundeformed
terrace surfacesand their associatedgravels. The similarity in
orientationof terracesT2 and T3 suggeststhat these two sur-
facesarenotsignificantly
different
in age.
Soils developedon the T2 terraceexhibit an unexpected
range
ofprofilemorphologies
(Figure6b).Someprofiles
exhibiteda highdegreeof development,
as evidenced
by sec-
ondary
clay
redistribution
(argillic
pallic
soils),
whereas
othersoilsareclearlylessdeveloped(younger)and indistinguishablefrom those of the uppermostdegradationterrace T3
(orthic brown soils). The variations in soil development suggest that the Waipara River was either aggrading during the
late Holocene or maintained a similar grade and elevation,
possibly
throughout
thelastglacial
maxima
andextending
intothemid-Holocene.
In either
case
theWaipara
River
formeda broadaggradational
fan (Canterbury
Surface)that
wascharacterized
bya series
ofback-filled
channels
ofvarying age that are accordantin height. If this is the case, then
age
estimates
fortheCanterbury
aggradation
surface
may
need
toberevised
downward.
•
•Oo
NI½OL ET AL.: ACTIVE FOLDING IN WAIPARA AREA OF NORTH CANTERBURY
1339
Uplift Rates
synclineimmediatelyeastof the marineterracesis offshore
andthe Waipara Syncline,which plungesgently to the SSW,
is spatiallyremovedfrom the uplift data. For the proposesof
The rates of uplift calculated on the Omihi Surface at
Glenray must be regardedas relative or local values, which
this work we have assumedthat the base of the fold-enveloping surfaceis approximatelyhorizontalbetweenWaipara and
the coastlineand use a range of minimum structure contour
valuesfrom-250 to -650 m. This range representsvalues in
provideinformationaboutdeformationassociatedwith folding. To derivean estimatefor the total uplift, a componentof
regionaluplift mustbe addedto theselocal values[cf. Ghani,
1978]. To relatethe Glenray datato the total uplift (i.e., to calculate the local and regional componentsof uplift), we have
independentlyanalyzeduplifted marineterraces(M1 - M3)
preservedat the Pacificcoasteastof the studyarea(Figure7).
Marine terrace elevation and age data are from Yousif
[1987]. The lowest marine terrace (M1) forming the coastal
plain to the eastof the studyarea, thoughnot directly dated,
wasassigned
anageof approximately
60 ka (6•80,stage3).
This age is based on the presenceof two overlying loess
units,separated
by a weak paleosol,andAokautereAsh within
the upperloessunit [Kohn, 1979].The ageof theM1 terraceis
consistentwith marine terrace data from Wanganui [Pillans,
1985, 1988] which similarly show the approximately 60-kyrold Rakaupiko marine terrace overlain by two loess units.
While there is little direct evidencefor the age of the two re-
mainingandmore elevatedmarineterraces(M2 and M3), ages
of c. 80 and 100 kyrs, respectively,are inferred using the sea
level data of Chappell and Shackleton [1986] and produce
uplift valueswhich are consistentwith the fold structurecontour data, as well as M1 uplift rates (Figures 7 and 8). In
Figure7 the locationsand inferredagesof the marine terraces
are plottedon a structurecontourmap constructedon the base
of the Pliocene(baseof the Kowai Formation),modified from
Wilson [1963]. A positive relationshipexistsbetweenuplift
ratesand structurecontourvalues. To quantify th',srelationship,the uplift rate is plotted againststructureccntourvalue
for eachlocation(Figure8). The uplift ratesappearto be intimatelyrelated to the underlyingfold geometries(see also,
Yousif [1987]). Wellman[1971], Ghani [1978] and $uggate
[1987] have documentedsimilar relationshipselsewherein
New Zealandand concluded that the spatial distribution of
uplift wasrelatedto the positionof foldspreservedin the underlyingTertiarystrata.Similarly, the Waipara data also suggestthat the incrementalfold-relateduplift is in accordwith
the total finite shortening,from which it is infen'edthat a significant componentof the measurablefolding developedduring the currentphaseof regionaldeformation(i.e., within the
Pleistocene).We concludethat, locally, the pattern of uplift
principallyreflectsgrowthof the underlyingfolds.
If it is assumedthat the range of marine ten'aceuplift rates
(1.36-2.16m/kyr) reflectsprincipallythe locationof the terracesrelativeto the folds, then the points of zero local uplift
(i.e., macroscopicsynclinehinge zones)and maximum total
uplift (i.e., macroscopic
anticlinehinge zones)can be located
and their values estimated.However, because not all structure
contourshave been sampled,the linear relationship determinedin Figure8 (line b) mustbe extrapolated
to the highest
and lowest structurecontourvalues, assuminga constant line
gradientto estimatethe completerange of uplift rates. The
maximum structurecontour estimateis derived by sampling
from the structurallyhighest point on the Cass Anticline
(structurecontourapproximately
890 m), while determining
the minimum structure contour value is more difficult, as the
the WaiparaSynclineadjacentto the CassAnticlinemaxima
and is usedto calculatethe estimatedminimum uplift. At the
intersection of these structure contour estimates and the sta-
tistical line of bestfit (Figure 8, line b), two uplift valuesare
inferred,a minimumof 0.68-1.20 + 0.2 m/kyr and a maximum
of 2.68 + 0.2 m/kyr.Theseuplift rate estimatesare an orderof
magnitudelarger than thosederivedby Wellman[1979] for
the sameregionbut are in broadagreementwith uplift rates
calculatedimmediatelyto the north [Ota et al., 1984] and
southwest[Cowan, 1992a]. The maximum value, 2.68 + 0.2
m/kyr, is inferredto approximatelyequalthe maximumuplift
rate for the region betweenWaipara and the coast(i.e., the
coastalranges).The minimum uplift rate is regardedas the
lowestlocaluplift rate.At the locationof minimumuplift the
local component
of uplift is approximatelyzero and the total
uplift is approximately
equalto the regionalcomponent.
The
regionalcomponent
of uplift at Waiparais thereforeinferred
to be between0.68 and 1.20 m/kyr, and local uplift may accountfor up to approximately55-75% of the total. However,
the component
of local uplift in the total is expectedto decline significantlywestwardas the regionaluplift associated
with theAlpine Fault and development
of the SouthernAlps
increases.
A comparison
of the uplift ratesfor the Omihi surfaceand
marineterraces(Figure8, linesa and b) illustratesa disparity
betweenthesetwo data setswith respectto the structurecon-
tourvalues;wherefor a given rangeof structurecontourvalues,the OmihiSurfaceuplift ratesaregreaterthanthoseof the
marineterraces.Sucha disparitywouldbe expectedif the uplift ratesof the Omihi Surfacewere high (relativeto the average rates of uplift for this part of the fold) during the
Holocene.Locally high uplift ratescouldhavebeen induced
by southwardpropagationof the Omihi Fault and a corre-
spondingwestwardmigrationof the range front into the
Waipara valley.
The presentanalysisof uplift hastwo importantimplicationsfor improvingourunderstanding
of the analysisand interpretationof uplift in areasof active folding.Previously,
the analysisof uplift in New Zealandhas mainly been confinedto areaswheregeomorphicmarkersurfaces,suchas marine and fluvial terracesof known age, are present.Although
extremelyuseful,this has obviouslimitationsbecausethese
surfaces
areusuallyof limitedextentandare rarelypreserved
in areasof greatestuplift. We suggestthat the analysispresentedin Figure8 mayformthebasisof a techniquedesigned
to produce
upliftratecontourmapsin areaswherelimiteduplift dataare available.The accuracyof suchuplift mapswould
dependon the errorsassociated
with the uplift and structure
contourestimates,the correlationcoefficientof the two vari-
ables,andproximityof themapto the raw data.It is possible
thatthesemapswill providea first approximation
to uplift
patterns
ona regional
scale(50-100km2),although
in areas
wheretemporalchanges
in the uplift rateshaveoccurred
the
mapcouldrequiremodification.
We mustalsoadda caution-
1340
NICOL
ET AL.: ACTIVE
Estimated
range
ofstructure
contour
values
-250
to-650
mN.
/'
/'
Waiv•a
.-
•-.
AREA OF NORTH
<5,5//,,.O'
ß/
07• O•./
.' / •,'z z7
./
•t•
•.•'
CANTERBURY
•'
• .••.
1.•,
• .-•
•
,'• •
170•-•'•",'•
•
•
IN WAIPARA
/' c)4•.•?'e•
ZI./
/o7
/'
A?,
•
/'
/'
6 / •A•.
_'
FOLDING
ß•
'•
Ss•ated struct•e
'•--17501......
..-'
..-''
__-2.
. ,'/.',,, ,•.--- l•ut--...
... -..
,,Yo•sff•./ ........ • .... ''' .--''' ..-'''
1.67
•
,' (198•
•-''
•Ul
...
...
•
'. 2.14
.
,,,,.-,•.
/,'z-....
...._IE•''
'•4Z•
UPLIFTEDFEATU
• ,,,,',,t,','•"
.........
7_
.....
'--Z:? 1.
/I[ .1.83
12+2
fluvial
terrace
,,/: ,,.... 75-....
- ka
,•_,' ,' b'.' ,' .'
TE•OTDALE
•x
/
/up• rate
(m/kyr)
/ • •,"',.' ,',,' •,.'',,' .__
•_
••, •/
• kyr- UpPed
,'•,'/
''. ....
•w•CL•'
•1.85c.M1
, .• •' .' •.'. t•KAIEA
•*?•7• '
--q 8•/
•1.67
•
marine
"•"-'
• •--'" •
"•'•
•1.93 c.
80
kyrterraces
•
/
,
. 1.67 1.50 1.43 •_.
N
•
•
••
-----•'
•
•,
ß •
',
........
0
I.. I
•
•
.
•
'-
-''
•
,
,,
•
•
'•
ß
• - - -. •1.58
•
_• •
M3
c.100kyr
SY•OLS
''".1.3•'''/'•
._IC OCEAN ,•/" .Reverse
Fold
•al
trace
fa•t
or
th•st
St•ct•econto•
Figure 7. A comparisonof the spatial distributionof uplift rates, derived from both fluvial and marine terraces,and fold locationsas indicatedby structurecontourvalues. The structurecontoursare constructedon
the baseof the Pliocene(baseof the Kowai Formation)and are modifiedfrom Wilson [1963]. Uplift ratesare
indicatedin metersper kyr next to eachupliftedterracefeature.See text (Uplift Ratessection)for furtherexplanation.
ary note concerninguplift rates derivedfrom areasof active
folding,particularlywherethe ratesof regionaluplift are relatively small. It is quite clear that, locally, uplift rates vary
considerably.In this study, uplift rates are inferred to range
from 0.68 to 2.68 m/kyr acrossthe active folds in the coastal
ranges.To derive a rate of uplift, locally, as would be the case
for most marine or fluvial terrace data, and to use this value in
a regional analysis of uplift could producemisleading resuits. Therefore
extreme
caution
should be exercised
when
comparingratesof uplift spatially without an understanding
of wherethe uplift estimateswere derivedrelative to the local
fold structures.
Timing and Strain Rates of Deformation
Althoughthereis stratigraphicevidencefor tectonicactivity in North Canterburyduring the Miocene and Pliocene
[e.g., Andrews, 1963; Wilson, 1963; McCulloch, 1981;
BrowneandField, 1985], it appearsunlikely that measurable
faulting and folding began prior to the early Pleistocene
[Nicol, 1992; Cowan, 1992a, b]. This early Pleistoceneage
representsthe upperage limit of the Kowai Formation,which
based on outcrop scale observationsis everywhere conformablewith the underlying formations at similar locations
with respectto the macroscopicfolds. It has been suggested
thereforethat the onset of measurabledeformationpostdates
deposition of much of the Kowai Formation [Nicol, 1992].
This formationis everywherepoorlydated,but upperparts of
the unit are inferredby severalworkers[Wilson, 1963; Gregg,
1964; Browne and Field, 1985] to be Nukumaruan
(approximately 1.4 o 2.4 m.y. old [see Pillans, 1990]).
Conservatively,this would place the onset of deformationat
an age youngerthan 2.4 m.y.
Data collectedduringthis study suggestthat the development of faults and folds may be as young as approximately
0.5-0.8 m.y. Two independentsetsof data supportthis conclusion.Firstly, by using the rates of shorteningderived at
the Fault Bank locality and extrapolatingtheseback in time
(assuminga constantrate of shortening),all of the deformation observed within the Kowai Formation
can be achieved in
NICOL
ET AL.'
ACTIVE
FOLDING
IN WAIPARA
AREA
OF NORTH
CANTERBURY
1341
u
•
6
--L•
o
o
o
•
o
o
ß
o
o
LG
LG
i
LG
t",,
i
c)
1342
NICOL
ET AL.:
ACTIVE
FOLDING
IN WAIPARA
approximately
0.62 + 0.10 m.y. (Table2). A secondestimateis
derived from the 1340 + 250 m structural elevation difference
betweenthe CassAnticline and Waipara Synclineand a differentialuplift rateof 1.74 + 0.5 m/kyr (2.68-0.94). By assuming
that the rate of differential uplift acrossthe folds was constant
for the duration of deformation, we estimate that folding was
initiatedbetween0.5 and 1.3 m.y. ago. Althoughboth calculations require assumptionsthat are likely to representoversimplifications,we suggestthat, collectively,thesedata provide a basis for first approximationof the onsetof deformation. We favor a first-orderage of 0.8 + 0.4 m.y. for the onset
of deformationin this part of North Canterbury.Similar ages
have been inferred for the initiation of rapid uplift of the
North Island axial ranges[e.g., Ghani, 1978; Beu et al., 1981]
and the West Coast, SouthIsland [Nathan et al., 1986]. These
agedatasuggestthat a dramaticincreasein the rates of shortening and uplift occurredalong much of the plate boundary
in New Zealand within the last 1 m.y.
Our age data are combined with shortening estimates
(approximately6%), measuredalong the base of the Kowai
Formation from the crosssectionin Figure 1 and structural
contoursin Figure 7, to derive strain rates acrossthe coastal
ranges.A shortening rate of approximately 0.8 + 0.4%/100
kyr is calculated,a value intermediatebetween the rates determined at the Fault Bank (5.57%/100 kyr) and Glenray
(0.15%/100 kyr) localities. About 1 km of absoluteshortening has occurredacrossthe coastalrangesat Waipara over a
distanceof 16-17 km, at an averagerate of approximately1.4
+ 0.6 m/kyr. This rate representsonly a small proportion of
the predictedplate motion vector of about 38 m/kyr [De Mets
et al., 1990], as is to expectedon the edgeof the plate boundary zone. However, immediately to the north and west of
Waipara, shortening of 10-20% is more common [Nicol,
1991], and assumingthat the deformationis not significantly
older, ratesof shorteningof 3-5 m/kyr may be more representative.
Influence
of Deformation
Stratigraphy
on Cover
Bed
and Geomorphology
Although folds appear locally to have developed in responseto faulting at depth, the predominanceof folds at the
surface and the way in which deformation is laterally distributed by folding have a profound influence on surfaceprocesses.Active folding plays an integral part in establishing
the distribution and morphology of both geomorphic surfaces and associatedcover beds of varying ages in North
Canterbury.
As tectonic uplift is closely related to folding, so, too,
must the spatial distribution of erosion and sedimentationbe
significantly influenced by folding. Throughout North
Canterbury,anticline hingesare marked by eroding ranges or
ridges, while synclinesare coincident with valleys or basins
that contain late Pleistoceneaggradationgravels. These gravels often form angular discordancesof 2-30 ø with Tertiary
bedrock(e.g., at the Fault Bank locality,Figure4), which suggeststhat the onsetof sedimentationpostdatedthe initiation
of folding and shortening.Therefore the gravels are necessarily younger than 0.9 + 0.4 m.y.; at the Fault bank locality,
AREA
OF NORTH
CANTERBURY
sedimentationis inferredto postdatethe onsetof deformation
by ab.out 0.5 m.y. Similarly,the coastalrangeseastof Waipara
must be older than the oldestvalley fill, which from the Fault
Bank locality appearsto be approximately 100 kyr. Where
observedbedrock/late Pleistocene gravel contactsinvariably
dip toward adjacent synclines, Quaternary cover beds are
thickest above the active synclinal fold hinges and thin toward the anticlinal hinges.Evidencefor this is provided by
the absenceof Quaternarydepositson or adjacentto the anticlines and by the absenceof bedrock along the banks of the
WaiparaRiver within eachof the synclines.The height of the
first terrace above the southbank of the Waipara River suggeststhat Quaternarygravel depositsreach a thicknessof at
least20-30 m. By extrapolatingthe dip of the bedrockstrath
surfaceto the syncline hinges it is estimated that, locally,
gravel thicknessesmay reach maximum values of 80-140 m
above synclinehinges(Figure 1, crosssection).
Geomorphicsurfacesexaminedduringthis study appearto
have formedwithin the last 100 kyr and are young relative to
the ageof onsetof deformation.Owing to active folding, fluvial surfacesof >10 kyr are mostoften preservedon the flanks
of anticlines. Local evidence for burial
of fluvial
sin'faces
closeto synclinesis providedat the Mound Section,wherea
loess-covered
surfacepassesbeneatha degradationsurfaceof
youngerage (Figure 6a). Conversely,late Pleistocenecover
beds and fluvial terraces are currently being uplifted and
erodedalong the Waipara rangefront.
Below the Omihi aggradationsurface,flights of degradation
terracesstepdownto the presentWaipara River floodplain.
The degradationsurfacesare generallycharacterized
by rudimentary stony soil profiles which strongly suggest late
Holoceneage. Two radiocarbondates,eachof about1 kyr B. P.
also suggestthat most of these degradationsurfacesare of
late Holoceneage.At Waiparathe Waipara River has downcut
approximately20 m in the last 1 kyr, which would requirean
averagedowncuttingrate of 2 m/100 years over this period.
The reasonfor this period of rapid downcuttingis not clear,
althoughit may be that river incisionwas triggeredby uplift
associatedwith the last fold growth event on the adjacentanticline. If this is so,thenthe questionremainsas to why there
is little evidenceof previous Holocene downcuttingevents
along this part of the Waipara River. Few data are presently
availableto addressthisproblem,and we cannotdiscountthe
possibility that close to the Waipara Synclinethe Waipara
River hasexperiencedperiodsof bothaggradationand degradationduringthe Holocene.Rapid late Holocenedowncutting
requires that despite continued deformationthe Waipara
River remained at or close to the level of the late Pleistocene
aggradationsurfacefor possiblyas long as 10 kyr. It is also
reasonableto speculatethat rapid downcuttingof the Waipara
River duringthe late Holocenemight havebeeninfluencedby
two narrow, structurallycontrolled gorges that confine the
river on either sideof the Waipara Syncline and control rates
of sediment flux.
Conclusions
Rates of folding in the Waipara area during the late
Quaternaryaremeasured
using tilted and uplifted fluvial terraces, marine terraces,and cover beds. Our data suggest that
NICOL
ET AL.: ACTIVE
FOLDING
IN WAIPARA
the spatial and temporal distribution of deformation is variable. The highest rates of deformation occur along the
Waipara range front, where bedrockdeformation is most intense.Shorteningrates reach up to 5.57 + 0.69%/kyr and average 0.8 ñ 0.4 %/ 100kyr acrossthe coastalranges.This average is indicative of an absolute shorteningrate of 1.4 + 0.6
m/kyr, which representsonly a small proportion of the predictedplate motion vector in this region but may increaseto
3-5 m/kyr immediately to the north and west. Uplift rates
rangefrom 0 to 1.83 m/kyr for a late last glacial fluvial terrace
and from 1.36 to 2.16 m/kyr for three marine terraces. Both
fluvial and marine terrace uplift rates vary in accordwith the
geometriesof folds in the bedrock, and the spatial pattern of
uplift directly reflects fold growth. In these circumstancesa
first approximationto the spatialpattern of uplift may be derived from the structure contour pattern of beds across the
folds. Fold-related differential uplift accounts for up to approximately 55-75% of the total uplift and has produced
References
Alloway,B. V., B. J.Pillans,A. S. Sandhu,and
J. A. Westgate,Revision of the marine
chronologyin the WanganuiBasin, New
Zealand, based on the isothermalplateau
fission-track dating of tephra horizons,
Sediment.Geol., 82, 299-310, 1993.
Alloway,B. V., D. J. Lowe,tL P. K. Chan,D.
Eden,and P. C. Froggatt,Stratigraphyand
chronology
of the Stenttephra,a c. 4000
year old distal silicictephrafrom Taupo
Volcanic Centre, New Zealand, N.Z.J.
Geol.Geophys.,37, 37-47, 1994.
Andrews,P. B., Stratigraphicnomenclature
of
the Omihi and Waikari Formations, North
Canterbury,
N.Z. J. Geol.Geophys.,6, 228256, 1963.
Berryman,K.R., Age, height,and deformation
of Holocene
marine
terraces
at Mahia
Peninsula,Hikurangi Subductionmargin,
New Zealand, Tectonics,12, 1347-1367,
1993a.
Berryman,K.R., Distribution,height, and
deformation
of
late
Pleistocene
marine
terraces at Mahia Peninsula, Hikurangi
Subduction margin, New
Zealand,
Tectonics,
12, 1365-1379,1993b.
Beu, A. G., G. H. Browne,and T. L. Grant-
Taylor,New Chlamysdelicatulalocalities
in the centralNorth Island and uplift in the
RuahineRange,N.Z. J. Geol.Geophys.,
24,
127-132, 1981.
Browne, G. H., and B.D.
Field, The
lithostratigraphy
of lateCretaceous
to early
Pleistocene rocks of North Canterbury,
New Zealand,N.Z. Geol. Surv.Rec., 6,
Wellington,1985.
Campbell, I. B., New occurrencesand
distributionof Kawakawa Tephra in South
Island, New Zealand, N.Z. J. Geol.
Geophys.,29, 425-435, 1986.
AREA OF NORTH
CANTERBURY
1343
folds with structural relief of about 1300 m. These folds have
formed over the last 0.8 + 0.4 m.y. since the onset of late
Cenozoic deformation in the Waipara region. The coastal
rangesand Waipara valley, with its associatedgravel deposits, owe their existence to folding and therefore must
postdatethe onsetof deformation.
Acimowledgments.Funding for this work was provided by the
EarthquakeResearchFoundation (A.N.), a Soil Science Department
and Lincoln University Postdoctoralfellowship (B.V.A.), and Lincoln
UniversityInternalResearchgrants(P.J.T.). We would like to thank the
Berry, Ensor,Forbes,Harris, Millist, andWatsonfamiliesfor cheerfully
allowing accessto their properties;Hugh Cowan and JocelynCampbell
for many stimulatingdiscussions
during the courseof this work; Dave
Manningand Paul Froggattfor electronmicroprobeanalyses;Bernard
Sp6rli, Brad Pillans,and Hugh Cowan for reviewing an earlier version
of this manuscript;and, finally, Kelvin Berryman and Tracy Vallier,
who reviewedthe final manuscript.
Campbell, J.K., and A. Nicol, Holocene
folding and rupture on the Bobys Creek
Fault and related Quaternary deformation
along the Waipara River, in Field Trip
Guides: Geological Society of New
Zealand and New Zealand Geophysical
Society 1992 Joint Annual Conference,
editedby J. K. Campbell,Misc. Publ. 63B,
pp. 137- 153, Geol. Soc. of N.Z.,
Christchurch,1992.
Chappell,J., and N.J. Shackleton, Oxygen
isotopesand sealevel,Nature324, 137-140,
1986.
Chinn, T. J. H., Use of rock weathering-rind
thicknessfor Holocene absolute age-dating
in New Zealand,Arct. Alp. Res., 13, 33-45,
1981.
Cowan, H. A., Structure, seismicity and
tectonicsof the Porter's Pass - Amberley
Fault Zone, Ph.D. thesis, Univ. of
Canterbury,Christchurch,
N.Z., 1992a.
Cowan, H. A., Quaternary expansionof the
New Zealand plate boundary zone into
North Canterbury,in ConferenceAbstracts:
GeologicalSociety of New Zealand and
New Zealand Geophysical Society 1992
JointAnnual Conference,edited by J. K.
Campbell,Misc. Publ. 63A, Geol. Soc. of
N.Z., Christchurch,1992b.
De Mets, C., 1L G. Gordon,D. F. Angus, and S.
Stein, Current plate motions,Geophys.J.
Int., 101,425-478, 1990.
Eden, D. N., A Late Quaternaryhistory of the
Awatere
and
Wairau
Valleys,
Marlborough, New Zealand, Ph.D. thesis,
Victoria Univ. of Wellington, Wellington,
N.Z., 1983.
Froggatt,P. C., and D. J. Lowe, A review of
late Quaternary silicic and some other
tephra formations from New Zealand:
Their
stratigraphy,
nomenclature,
distribution,volume and age, N.Z J. Geol.
Geophys.,33, 89-109, 1990.
Ghani, M. A., Late Cenozoicvertical crustal
movements in the southern North Island,
New Zealand, N.Z.J. Geol. Geophys.,21,
117-125, 1978.
Gregg, D. R., Sheet 18, Hunmui (lst ed.)
Geological map of New Zealand, scale
1:250000, Dep. Sci. and Ind. Res.,
Wellington,N.Z., 1964.
Harris, M.H., Canterbury Gravels at Omihi
Stream, Waipara, Diploma of Sci. thesis,
Univ. of Otago,Dunedin,N.Z., 1982.
Hewitt, A.E., New Zealand soil classification,
LandResour.Sci.Rep., 19, 133 pp., Dep. of
Sci. and Ind. Res.,Wellington,N. Z., 1992.
Kirkaldy, P. H. S., and F. G. Thomas,Final
report on a seismic survey in the
CanterburyPlains area of New Zealand,
Geol./Geophys.
Rep.27, BPShellandTodd
Pet.Dev. Ltd., Wellington,N.Z., 1963.
Knuepfer,P. L. K., Estimatingages of late
Quaternarystreamterracesfrom analysis
of weatheringrinds and soils, Geol. Soc.
Am. Bull., 100, 1224-1236, 1988.
Kohn, B. P., Identificationand significanceof
a late Pleistocenetephra in the Canterbury
District, South Island, New Zealand, Quat.
Res., 11, 78-92, 1979.
Lewis, K. B., Growth rate of folds using tilted
wave-planed surfaces:
Coast and
continental shelf, Hawke's Bay, New
Zealand, Bull. R. Soc. N.Z., 9, 225-231,
1971.
Maxwell, P.A., Structural geology and preQuaternary stratigraphyof the Kaiwara
District, North Canterbury, New Zealand,
M. Sc. thesis, Univ. of Canterbury,
Christchurch,N.Z., 1964.
McCulloch, B., Geology of the Mt. Brown
Beds, M.Sc. thesis, Univ. of Canterbury,
Christchurch,N.Z., 1981.
Mew, G., J. L. Hunt, P. C. Froggatt, D. N.
Eden, andR. J. Jackson,An occurrenceof
1344
NICOL
ET AL.:
ACTIVE
Kawakawa Tephra from the grey River
valley, SouthIsland, New Zealand,N.Z J.
Geol. Geophys.,29, 315-322, 1986.
Milne, J. D. G., Upper Quaternarygeology of
the Rangitikeidrainagebasin, North Island,
New Zealand, Ph.D. thesis, Victoria Univ.
of Wellington,Wellington,N.Z., 1973.
Nathan, S., H. J. Anderson,R. A. Cook, R. H.
Herzer, R.H. Hoskins, J. I. Raine, and D.
Smale,
Cretaceous and
Cenozoic
sedimentarybasins of the West Coast
region,SouthIsland, New Zealand, in New
ZealandGeologicalSurveyBasinStudies,1,
Department of Scientific and Industrial
Research,Wellington, New Zealand, 1986.
Nicol, A., Structuralstyles and kinematicsof
deformation on the edge of the New
Zealandplate boundaryzone,mid-Waipara
region, North Canterbury, Ph.D. thesis,
Univ. of Canterbury, Canterbury, N.Z.,
1991.
Nicol, A., Tectonic structuresdeveloped in
Oligocenelimestones:Implicationsfor New
Zealand plate boundary deformation in
North Canterbury,N.Z. J. Geol. Geophys.,
$5, 353-362, 1992.
Ota, Y., T. Yoshikawa, N. Iso, A. Okada, and
N. Yonekura, Marine terraces of the
Conwaycoast,South Island, New Zealand,
N.Z J. Geol. Geophys.,27, 313-325, 1984.
Palmer, A. S., Kawakawa Tephra in
Wairarapa, New Zealand, and its use for
correlatingOhakea Loess, N.Z.J. Geol.
Geophys.,25, 305-315, 1982.
Pillaris, B. J., Southwest North Island
palcoenvironments150,000 years B.P. to
present, in Proceedings of the 2rid
CLIMANZ Conference, edited by B. J.
FOLDING
IN WAIPARA
AREA
OF NORTH
Pillans,Publ. $1, pp. 37-43, Geol. Dep.,
Victoria Univ. of Wellington, Wellington,
N.Z., 1985.
Pillaris,B. J., Loess chronologyin Wanganui
Basin, New Zealand, in Loess: Its
Distribution,Geology and Soils, edited by
D. N. Eden and R. J. Furkert, Proceedings
of the Intemational Symposiumon Loess,
New Zealand, 14-21 February 1988, pp.
175-191, A. A. Ballcema, Rotterdam,
Netherlands, 1988.
Pillaris, B. J., New Zealand Quatemary
stratigraphy:
An overview,Quat.Sci. Rev.,
10, 405-418, 1990.
Pillans,B. J., M. S. McGlone, A. S. Palmer, D.
Mildenhall, B. V. Alloway, and G. Berger,
The last glacial maximum in central and
southem North Island, New Zealand: A
palcoenvironmentalreconstructionusing
Kawakawa Tephra Formation as a
chronostratigraphic
marker, Palaeogeogr.,
palaeoclimatol.,palaeoecol.,101, 283-304,
1993.
Singh,L. J., Uplift and tilting of the Oterei
coast,Wairarapa,New Zealand,duringthe
last ten thousandyears, in Recent Crustal
Movements,editedby B. W. Collinsand R.
Frazer, Bull. R. Soc.N.Z., 9, 25-30, 1971.
Suggate, R.P., Active folding in North
Westland, New Zealand, N.Z J. Geol.
Geophys.,$0, 169-174, 1987.
Wellman,H. W., Holocenetiring anduplift on
the Glenbum coast, Wairarapa, New
Zealand, in Recent Crustal Movements,
editedby B. W. Collinsand R. Frazer,Bull.
R. Soc.N.Z, 9, 211-215, 1971.
Wellman, H. W., An uplift map for the South
Island of New Zealand, and a model for
CANTERBURY
uplift of the SouthemAlps, in Origin of the
SouthemAlps, editedby R. I. Walcott and
M. M. Cresswell,Bull. R. Soc. N.Z., 18, 1320, 1979.
Wilson, C. N., V. R. Switsur, and A. P. Ward,
A new date for Oruanui (Wairakei)
eruption,New Zealand, Geol. Mag., 125,
297-300, 1988.
Wilson, D. D., The northern continuationof
the Canterbury Plains, Canterbury, New
Zealand, N.Z.J. Sci. Technol.,$7, 126131, 1955.
Wilson, D. D., The geology of the Waipara
Subdivision(Amberleyand Motunau sheets
S68 and S69), N.Z. Geol. Surv. Bull., 64,
122 pp., 1963.
Yousif, H.M.S., The applications of remote
sensingto geomorphologicalneotectonic
mappingin North Canterbury,Ph.D. thesis,
Univ. of Canterbury, Canterbury, N.Z.,
1987.
B. V. Alloway, Department of Geology,
Auckland University,Tamaki Campus,Private
Bag 92019, Auckland, New Zealand. (e-mail:
b.alloway•auckland. ac.nz)
A. Nicol, Earth Sciences Department,
University of Liverpool, Liverpool, England
L69 3BX.
P. J. Tonkin, Departmentof Soil Science,
Lincoln University, PO Box 84, Canterbury,
New Zealand.
(Received December 16, 1993;
revisedMay 12, 1994;
acceptedMay 24, 1994.)