Antepenultimate glacial to last glacial deposits in southern Wairarapa, New Zealand

Download Antepenultimate glacial to last glacial deposits in southern Wairarapa, New Zealand

Post on 09-Feb-2017

216 views

Category:

Documents

3 download

TRANSCRIPT

This article was downloaded by: [Washington State University Libraries ]On: 16 October 2014, At: 00:53Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UKJournal of the Royal Society of NewZealandPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/tnzr20Antepenultimate glacial to last glacialdeposits in southern Wairarapa, NewZealandC. G. Vucetich a b , P. Vella a c & P. N. Warnes a da Geology Dept. , Victoria University of Wellingtonb 17 Ruru Street, Waikanaec 1 Vella Street, Titahi Bayd Institute of Geological and Nuclear Sciences , Lower HuttPublished online: 30 Mar 2010.To cite this article: C. G. Vucetich , P. Vella & P. N. Warnes (1996) Antepenultimate glacial to lastglacial deposits in southern Wairarapa, New Zealand, Journal of the Royal Society of New Zealand,26:4, 469-482, DOI: 10.1080/03014223.1996.9517521To link to this article: http://dx.doi.org/10.1080/03014223.1996.9517521PLEASE SCROLL DOWN FOR ARTICLETaylor & Francis makes every effort to ensure the accuracy of all the information (theContent) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &http://www.tandfonline.com/loi/tnzr20http://www.tandfonline.com/action/showCitFormats?doi=10.1080/03014223.1996.9517521http://dx.doi.org/10.1080/03014223.1996.9517521Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditionsDownloaded by [Washington State University Libraries ] at 00:53 16 October 2014 http://www.tandfonline.com/page/terms-and-conditionshttp://www.tandfonline.com/page/terms-and-conditions Journal of The Royal Society of New Zealand,Volume 26, Number 4, December 1996, pp 469482Antepenultimate glacial to last glacial deposits in southernWairarapa, New ZealandC. G. Vucetich, P. Vella and P. N. Warnes*In southern Wairarapa valley, north of the marine terraces adjacent to Palliser Bay,three fluvial aggradation gravel terraces with small vertical separation, at successivelylower levels, represent the Porewa, Rata and Ohakea stadials of the Last Glacial Stage.Their correlative loess deposits and interbedded Kawakawa and Middle TongariroTephras mantle much of the lower part of the valley and higher areas to the east,providing stratigraphic control on underlying surfaces of pre-Ohakean, pre-Ratan andPre-Porewan ages. A slightly higher terrace represents the Last Interglacial Stage. It hasa cover of Francis Line Formation, predominantly fine-grained fluvial overbank depositsc.5-7 m thick, locally thicker lacustrine deposits, with overlying Porewa, Rata andOhakea Loess with their interbedded tephras. The eastern limit of Francis Line Formationis defined approximately, and farther east the Last Interglacial Stage is represented bya paleosol, usually strong red in colour, and at places developed in thin wind-blownsand. Ahiaruhe Formation underlies Last Interglacial deposits unconformably, consistsmainly of alluvial gravel, but includes relatively thin loess and fresh-water silt and sandlayers, Mount Curl Tephra and a strong brown paleosol above the tephra. It is consideredto represent the Penultimate Glaciation, Penultimate Interglacial, and AntepenultimateGlaciation, and was deposited in a regime of tectonic subsidence.Keywords: river terraces; loesses; tephras; paleosols; eustatic sea-levels; tectonic deformationINTRODUCTIONVella (1963) mapped three fluvial aggradation terraces in Wairarapa Valley and correlatedthem from highest to lowest with the first, second and third stadials of the Last Glacial Stage.Palmer and Vucetich (1989) described cover-bed sequences comprising Porewa (First Stadial),Rata (Second Stadial) and Ohakea (Third Stadial) loesses overlying Last Interglacial Stagedeposits. Each loess layer is presumed to be derived from dust that was blown mainly by theprevalent north-westerly wind off the surface represented by its corresponding terrace whenthe terrace was forming as a broad aggrading fluvial flood-plain with little vegetation cover.Between Carterton and the Ruamahanga River, Warnes (1989, 1992) recognised fourterraces. The younger three terraces correlate with the three described by Vella (1963). Theoriginal tread of the oldest (fourth) terrace is the top of Francis Line Formation (Warnes1992), consisting of mainly fine-grained fresh-water deposits with some lenses of sand andfine gravel. It now has a mantle ranging up to five metres thick containing the three loessunits derived from the Porewa, Rata and Ohakea flood-plains, and is correlated with the LastInterglacial Stage. New observations (Appendix Table 2; Fig. 1) define the approximateeastern margin of the Francis Line Formation.* All formerly at Geology Dept., Victoria University of Wellington. C.G.V. now 17 Ruru Street,Waikanae; P.V. 1 Vella Street, Titahi Bay; P.N.W. Institute of Geological and Nuclear Sciences, LowerHutt.Downloaded by [Washington State University Libraries ] at 00:53 16 October 2014 470 Journal of The Royal Society of New Zealand, Volume 26, 1996wmw/Fig. 1 Map of southern Wairarapa Valley showing sampling sites listed in Appendix Tables 1 and 2.Heavy dash-dotted line shows approximate eastern edge of Last Interglacial fine-grained fluvial andlacustrine deposits (Francis Line Formation) and estuarine sandy mudstone (site J). Remnants of the6,500 y (BP). eastern shoreline of Lake Wairarapa are shown for comparison.Mount Curl Tephra*, loess, and a prominent paleosol here correlated with the PenultimateInterglacial Stage are aeolian deposits buried within fluvial aggradation gravels of theAhiaruhe Formation. The aeolian origin was not recognised in the original description of theformation (Collen & Vella 1984). The tectonic environment of deposition and status of theAhiaruhe Formation are discussed below.*Mount Curl Tephra is sometimes referred to as Rangitawa Tephra. The holostratotype of RangitawaPumice described by Te Punga (1952, 1959) is a 6 m thick bed that is certainly mostly reworked tephricmaterial and has never had any primary airfall tephra described within it by Te Punga or anybody else.We examined it and could find no primary airfall tephra. What has been called Rangitawa "Tephra" ismore correctly referred to as Mount Curl Tephra, with its plainly primary airfall holostratotype atMount Curl Road (Milne 1973).Downloaded by [Washington State University Libraries ] at 00:53 16 October 2014 Vucetich Glacial deposits in southern Wairarapa 471LAST GLACIAL STAGE DEPOSITSThe three names Ohakea, Rata and Porewa were first applied to fluvial aggradation gravelformations, each well preserved in terrace treads ascending in order of increasing age atRangitikei Valley (Te Punga 1952; Milne 1973). Each tread had its origin as a flood-plainfrom which channel silts were blown by wind to be deposited on higher surfaces as a loessmantle. The phases of alluvial aggradation and accompanying loess accumulation date to theLast Glaciation cold climate phases (stadials) when forest was sparse. Phases of riverdowncutting date to the intervening warmer climate phases, with narrow flood-plains, morevegetation cover, and little loess generation. Because of continuous uplift of the Rangitikeidrainage basin, the relation of alluvial aggradation to rapid loess deposition was easy torecognise; e.g. Ohakea Gravel is the stratigraphic equivalent of Ohakea Loess which formsthe top of the mantle on higher terrace gravels. The terrace gravels, derived from RuahineRange Mesozoic (Torlesse) greywacke, were partially protected from weathering by theloess mantle. Each loess unit itself was subject to soil weathering both whilst it was rapidlyaccumulating and during its following interval of negligible accumulation. Differing weatheringmodes resulting from local factors, chiefly differences of climate and of the freedom ofsubsurface drainage, are reflected by five loess facies recognised in New Zealand: aerieloams, aerie earths, aerie dense silts, aquic dense silts and clay silts (Milne 1988).The three Last Glacial Stage fluvial aggradation terraces between Carterton and theRuamahanga River show the same genetic relationship to loess deposits as that demonstratedat Rangitikei by Milne (1973) and in northern Wairarapa by Vella et al. (1988). The mostthorough descriptions of southern Wairarapa Last Glacial Stage loess sequences are those ofPalmer & Vucetich (1989), based on continuous cores taken using a truck-mounted drill.Three of the cored sites, H and I (Fig. 1), and another to the south near Lake Ferry on thePalliser Bay coast, are discussed below. All three overlie deposits of Last Interglacial age andcontain three distinct loess units separated by paleosols and correlated from core to core bylithological and soil weathering characters and by two tephric horizons. Other loess sequences,reported by Warnes (1989, 1992) and in this article (fig. 1; Appendix Tables 1, 2 & 3), havebeen determined by Jarrett augering.Almost invariably uppermost in the cover-beds is Ohakea Loess positively identifiedbecause it contains Kawakawa Tephra (Aokautere Ash of Cowie 1964), carbon14 dated as22 600 y old (Wilson et al. 1988). Kawakawa Tephra is rarely found as a macroscopic layerin southern Wairarapa. It is exposed as a 510 mm layer in a cutting on the Martinborough-Greytown Road adjacent to core site H, and Palmer and Vucetich identified it in the core1.5 m below the ground surface, two-thirds of the way down the thickness of Ohakea Loess.Warnes (1989) detected it microscopically as concentrations of glass shards in loess sandfractions from many sites, identifying it by glass chemistry using electron microprobeanalyses. The depth of the tephra below the top of the Ohakea Loess varies from place toplace. At Riverside, 4.5 km north of Martinborough at the top of a cliff on the east side of theRuamahanga River, the Ohakea Loess is 7 m thick and the Kawakawa pumiceous deposit isa prominent layer at half that depth. The extraordinary thicknesses of the loess and pumicethere are attributed to strong updrafts from the Ruamahanga flood-plain during north-westerly gales. In contrast the tephra is only 0.5 m deep or shallower in the area betweenCarterton and the the Ruamahanga River, (e.g. at site D, Appendix Table 3). The differencesin depth of the Kawakawa Tephra reflect differences in the amount of loess received at thedifferent sites. Each of the three loess units tends to vary in thickness from site to site (seeAppendix Table 3 for examples).The drilled cores (Palmer & Vucetich 1989) define lateral facies difference in the loesswithin southern Wairarapa, which need to be appreciated when correlating stadial units fromsite to site. The Lake Ferry core sampled a mantle of loess overlying freely draining gravel ofa Last Interglacial marine terrace mapped by Ghani (1978). The mantle contains three loessunits which by superposition are correlated with the Ohakea, Rata and Porewa Loesses of theLast Glaciation. A prominent reddish brown layer in the sequence, the Lake Ferry Paleosol,Downloaded by [Washington State University Libraries ] at 00:53 16 October 2014 472 Journal of The Royal Society of New Zealand, Volume 26, 1996exposed near the core site, is a key marker bed between Rata Loess above and Porewa Loessbelow. The reddish brown colour has arisen from dominantly oxidising soil weathering of aniron-rich andesitic ash accretion (Middle Tongariro Ashes, Milne 1973) and contrasts withthe yellow-brown colours of the low iron quartz-feldspar dominant loess units derived fromTorlesse greywacke. Polygonal columnar cracks form a conspicuous veined pattern in allthree loess units. The cracks record former seasonal wet-dry cycles in Rata and Porewaloesses and ongoing cycles in the present ground soil and Ohakea Loess. The cracks in theRata and Porewa Loesses might be still active but more probably are entirely sub-fossil.The Bidwill core (site H, Fig. 1) contains the three loess units overlying Last Interglacialpeat and impermeable grey and bluish-grey water-laid clay which impede drainage andremain wet the year round. In the adjacent road cutting polygonal columns are conspicuous inOhakea Loess but weakly developed in Rata Loess. Palmer and Vucetich (1989) describeRata Loess in the core as firm silty clay loam, very plastic when wet, with a dull yellowishorange base colour and profuse orange and light grey mottles. It is distinguished from theOhakea Loess above by its texture, stronger orange colour and blocky structure. The LakeFerry Paleosol separating Rata from Porewa Loess is not distinguished by the strong colourseen at Lake Ferry. The Porewa Loess is massive and totally mottled grey and orange, incontrast to its blocky structure and yellow brown colour at Lake Ferry.Porewa and Rata Loess reported by Warnes (1989) and at several new sites listed here(Appendix Table 2) also overlie Last Interglacial clayey fresh-water deposits (Francis LineFormation, Fig. 2) which impede subsurface drainage, and resemble the facies at the Bidwillcore site.LAST INTERGLACIAL DEPOSITSThe Porewa Loess is here taken to represent the First Stadial of the Last Glaciation, as it wasby Milne (1973) correlated with oxygen isotope Stage 4, with its base slightly younger thanthe marine bench on which it rests at Lake Ferry, inferred to be 80 000 y old by Ghani (1978).Berger et al. (1992), using the thermoluminescence method, estimated an age of 85 000 +15 000 y for a sample from the Lake Ferry core 40 cm above the top of the gravel. All of thesites where Porewa Loess has been recognised in southern Wairarapa inland from the LastInterglacial marine terraces adjacent to Palliser bay (Ghani 1978) are marked on Fig. 1.Lettered sites were reported by previous authors (Appendix Table 1) and numbered sites arereported here for the first time (Appendix Table 2). Sites Al and B east of Carterton, J nearthe southern edge of the map, and 6 at White Rock Road near Martinborough, are usefulexposed outcrops. Sites 9 and 11 are poorly exposed.At nearly all sites west of the heavy dash-dotted line (Fig. 1) Porewa Loess is conformablyunderlain by predominantly fine-grained fresh-water deposits of the Francis Line Formation(Warnes 1992) which is correlated on the grounds of stratigraphic superposition with the LastInterglacial Stage. The Francis Line deposits include clay and silty clay rarely interrupted byprobably lensoid thin sand and fine gravel intervals. The clays are typically grey, greenishgrey or blue grey but at places are oxidised to shades of yellow or orange. Sand and gravelintervals arc usually oxidised to yellow brown, rarely red brown. At some sites the clayscontain abundant rhizomorphs and at other sites carbonised rootlets suggesting anoxicswampy environments of deposition. The Bidwill core site (H, Fig. 1) is the only place wherepeat has been found. In contrast, at site 4, 2 km south of site H, near the eastern edge ofBidwill Ridge, Porewa Loess rests on an iron-enriched red brown paleosol overlying ayellow brown alluvial bar gravel.Warnes (1992) reported exposed thicknesses of Francis Line Formation >8.5 m at FrancisLine Road east of Carterton and >3.0 m at the former Carterton Brickworks quarry on theeastern edge of Carterton. No augered or cored section has penetrated much more than 1.0 mof it. Palmer & Vucetich (1989) included reports on fossil pollen and diatoms by M.S.McGlone and M.A. Harper that respectively indicate deteriorating climate up-section, andfresh-water deposition. Warnes (1992) included reports by W.A. McLea and M.A. HarperDownloaded by [Washington State University Libraries ] at 00:53 16 October 2014 IT; _ JVucetich Glacial deposits in southern Wairarapa 473 SouthPAHIATUAREFUSE TIPWOODLANDSROAD SITE DFOREMAN-JURY ROADBIDWILL RIDGECORE, SITE HohXXXHTraTTTGreenhills=Harton loessU. Griffins Rd TephraFlat Top=Burnand loessL. Griffins Rd tephraRidge Rd 1 loessPenult. Intergl.paleosolMount Curl TephraRidge Road 2=Aldworth LoessRidge Road 3 Loesswhite clayPahiatua Terracegravel15m omittedupperAhiaruhegravellowerAhiaruhegravel& silt55m omittedgr. siltpeatnot bottomedby core.Penetrometershowed ?5mAhiaruhegravelexposednearbyMesozoicgreywackeWHITE ROCKROAD SITE 5LAKE FERRYCORETTTLilLakeFerryPaleosol84 OOOymarineterraceFrancisLine Fmtn26m omittedAhiaruhegravelLegendloessIpaleosolBrhyolitictephraclay silt& sandMesozoicgreywackeFig. 2 Correlations of late Quaternary deposits in representative southern Wairarapa sections and thenorthern Wairarapa reference section at Pahiatua refuse tip (Vella et al. 1988). Key marker beds areKawakawa Tephra and Mount Curl Tephra. Foreman-Jury Road section from Collen and Vella (1984);Lake Ferry core from Palmer (1982). Localities of other sections detailed in Appendix.indicating sparse abraded robust fossil pollen grains and fresh-water diatoms inferred to havebeen transported by running water.Most of Francis Line Formation appears to be river overbank flood deposits, much of itlaid down in swampy situations. The Francis Line Road and Carterton Brickworks depositsmay represent small shallow lakes. The 0.6 m sampled at the base of the Bidwill core (site H)might have been deposited in a shallow pond like those that now exist on the eastern side ofLake Wairarapa. A farmer's description of a water bore suggests that there is a fairly thick(30 m?) lake deposit south of Martinborough (Fig. 3 A). The consistent stratigraphic positionof Francis Line Formation, conformably underlying Porewa Loess and unconformablyDownloaded by [Washington State University Libraries ] at 00:53 16 October 2014 474 Journal of The Royal Society of New Zealand, Volume 26, 1996A A " N N E2001 uIDicS 200I 150I ioo-MetSObMetreso, Martinborough-Lake Ferry RdPresent sea-level -Vertical scale - horizontal scale tSame cross-section as above butvh x10 vertical exaggerationOhakean anRatan Vloesses.Porewa; i n - i/ ^ NSITE 5 / cliff $- - - - ' 'vet?' Present saa-bm0 1)" WNW. Holocene? Last Interglacialsurface surfaceSITEH SITE 41(S S E - - B " CHarris Trig 164 mDry River Fault / Huangarua(dip uncertain) SynclineSITE 6Gravel older than Last IG; eroded on Harris Ridge;overlain by Last IG paleosoland wind-blown sand? 3 _HolocenesurfaceRuamahanga R.Present^ verticalscate. horizontal0 1 ? 3 _Ohakean iRatan \- bePorewan Jrecent swamp^ J ^ - * ? ?Jury Ridge^ ^ ^ ^ ^ ^scRuakokoputuna R.=-- r^AI- ~VPliocene and lowerQuaternary marine strata* 7Sinll P H f 0 h'"' ?For older strata see truescale cross-section aboveE S E -PonatahtValleyMesozoic greywacke Upper CenozoicSubsurface structure not knownFig. 3 A. Vertical geological cross-sections along White Rock Road (line A-B-C, fig. 1). The NWend (A) is lkm SW from Martinborough. The upper cross-section is drawn at true scale (vertical =horizontal) to show real topography and underlying structure of marine strata (mainly from Lamb andVella, 1987). The lower cross-section, along the same line, has a xlO vertical scale exaggeration toshow relations of the late Quaternary deposits. B. True scale topographic profile along line D-E (fig.1), 8 km N of Martinborough, showing tilted down to WNW Last Interglacial terrace on Bidwill Ridgeat left, and tectonically rising older rocks to right.overlying Ahiaruhe Formation and older rocks, confirms its correlation with the LastInterglacial Stage by Warnes (1992).Bidwill Ridge (Fig. 4A; left part of Fig. 3 A) is a large uplifted and tilted remnant of theLast Interglacial flood-plain. The mean tilt (0.95 degrees down to WNW) indicates a tilt rateof 8 degrees per million years, similar to that shown by middle Quaternary strata in theeastern flank of the Huangarua Syncline adjacent to White Rock Road 14 km to the south(Lamb & Vella 1987). Between Carterton and the Ruamahanga River Warnes (1992)recognised folding of the Last Interglacial Terrace with SSW-NNE axes over anticlinesdefined in seismic reflection profiles (Cape et al. 1990).Along the axial trend of the folding the average height of the Last Interglacial flood-plaindecreases toward SSW at about the same gradient as that of the adjacent present RuamahangaRiver flood-plain, suggesting a depositional slope. The eastern limit of Francis Line Formation(heavy dash-dotted line, Fig. 1) marks the approximate eastern edge of the flood-plain. Theedge has been located most accurately at White Rock Road, south of Martinborough. Site 7(Fig. 3A) has the three Last Glacial loesses overlying a Last Interglacial strong red paleosoldeveloped on thin sand thought to be wind blown, underlain by yellow brown gravel ofunknown age. Site 6, 3 km to the north, exposed in a quarry beside Harris Trig, has the threeloess layers and a similar paleosol thought to be developed in thin sand unconformablyoverlying eroded red brown Ahiaruhe Formation alluvial pebble-cobble gravel (Fig. 4B).Downloaded by [Washington State University Libraries ] at 00:53 16 October 2014 Vucetich Glacial deposits in southern Wairarapa 475Site 5 a further 1.5 km to the north (Fig. 3A) has the three loess layers overlying a lessconspicuous yellow brown paleosol formed on top of Last Interglacial light brown greyweathered fresh-water clay with rhizomorphs (Francis Line Formation); 100 m NW is thefarmer's water bore mentioned above with c.30 m of lacustrine blue grey silt. Site 12, 1.2 kmSSW from site 5 (Fig. 1) has a thick three loess sequence, Ohakea 3.1 m, Rata 1.1m andPorewa >1.4 m, not bottomed. Nearby the Porewa Loess bottoms on gravel with no paleosol,considered to be a thin lens or wedge (shown diagramatically, Fig. 3A) derived by erosion ofAhiaruhe Gravel on the north-west slope of Harris Ridge nearby to the south-east. A low cliff800 m south of site 5 (Fig. 3 A), mantled by thick colluvium and undifferentiated loess, hasbeen cut into Ahiaruhe gravel at the north-west edge of Harris Ridge and is inferred to markthe east shore of a lake with a large enough fetch to develop moderately erosive waves.The Holocene Lake Wairarapa and Lake Onoke developed similar low cliffs on theireastern sides when at their maximum width >50006000y ago. They are now much reducedin size with their original eastern area infilled by Holocene sediments slightly raised abovethe present lake levels by tectonic tilting and perhaps a component of slight sea-level (base-level) fall since the 6500y BP culmination of the post-glacial sea-level rise. These Holocenelakes, their now emergent lacustrine sediments and overlying aeolian sands, and the southernc.20 km of the Ruamahanga River flood-plain occupy a nearly infilled Last Glacial Stageriver valley system that was cut down to a lower base-level than the present one and wasdrowned by the post-glacial rise of sea-level. They provide an analogue for interpreting theLast Interglacial deposits which were laid down after drowning of the Penultimate Glaciationtopography by the Last Interglacial high sea-level, which probably reached a maximum atleast as high as present sea-level and possibly 6 m higher (Kennett 1982, p. 272).The farthest south observation site shown (J, Fig.1) is an exposure of blue grey sandymudstone containing abundant calcareous fossils of two species only, the bivalve Austrovenusstutchburyi (common cockle) and the foraminifer Ammonia beccarii. The sediment andfossils there indicate a low energy and low salinity inner estuarine environment of deposition,are about 40 m above sea-level, consistent with the local uplift rate of Last Interglacialmarine terraces (Ghani 1978) and indicate a Last Interglacial analogue of the present LakeOnoke, which in turn indicates a southern limit north of it for the fresh-water Francis LineFormation. The inferred broad and +30 m thick lake deposit south of Martinborough andwest of Harris Ridge would have resulted from drowning of the Penultimate Glacial StageRuamahanga Valley (Fig. 3 A) and thus would be an analogue of the Holocene lake depositsof Lake Wairarapa. The present relatively narrow flood-plain of the Ruamahanga River northof Martinborough can scarcely be compared to the Last Interglacial flood-plain which was atleast two or three times wider at places, with much of its western limit still not determined.Tectonic tilting rates seem not to have increased during the last million years, but tectonicuplift of Bidwill Ridge has confined the Ruamahanga River to a gorge in the brief time sincethe Last Interglacial Stage.AHIARUHE FORMATIONAhiaruhe Formation (Collen & Vella 1984) contains yellow to red brown alluvial gravelswith pebble to cobble subrounded greywacke clasts, fresh-water green-blue-grey clayey siltlayers, infrequent metre scale discontinuous yellow brown well sorted vertically jointed siltlayers (some of them certainly loess), and a medium brown paleosol overlying Mount CurlTephra (Fig. 2). Near its type locality the formation is partially exposed and at least 64 mthick. It unconformably overlies Mesozoic greywacke on the south side of RuamahangaRiver at the north end of Jury Ridge, upper Pliocene limestone 2 km east of the north end ofJury Ridge, and Pliocene? marine sandy mudstone with fossil Radiolaria at White Rock Roadadjacent to Harris Trig (Fig. 1). It is geographically separate and lithologically different fromthe middle Quaternary Te Muna Formation, and regarded as younger because it containsMount Curl Tephra and none of the tephras (all older) found in Te Muna Formation. BecauseDownloaded by [Washington State University Libraries ] at 00:53 16 October 2014 476 Journal of The Royal Society of New Zealand, Volume 26, 1996 ''"*=Fig. 4 Last Interglacial surface; Ahiaruhe Formation. A: View south-west across Last Interglacialsurface from Martinborough-Greytown Road on Bidwill Ridge 4 km north of junction of Featherston-Martinborough Road. The surface is gently undulating owing to varying thickness of loess cover (up to5 m). B, Gravel quarry west side of White Rock Road, 500 m N.E. of Trig Harris, S27 166915.Approximately 10 m of weakly cemented iron oxide coated trough cross-bedded sub-rounded pebble-cobble alluvial gravel of Ahiaruhe Formation; lighter coloured layer at top overlies unconformably andincludes in upward succession Last Inter-glacial thin sand and strong red paleosol, Porewa Loess, LakeFerry Paleosol, Rata Loess and Ohakea Loess. At roadside 50 m around left of photo Ahiaruhe gravelunconformably overlies Pliocene marine sandy mudstone.its subaerial tephra, paleosol and loess are buried within thick alluvial deposits it is inferredto have been laid down during a regime of tectonic subsidence.Most of the formation is still buried, and outcrop exposures provide only incompletestratigraphic sequences (Fig. 2). Three large exposures are: 1) Inaccessible cliff east side ofRuamahanga River, south side of bridge on Carterton-Ponatahi Road, c.20 m of yellowDownloaded by [Washington State University Libraries ] at 00:53 16 October 2014 Vucetich - Glacial deposits in southern Wairarapa 477pal-eosotbloessFig. 4 C, South side of Highway 53 (Featherston-Martinborough) at junction of Wards Line RoadS27 128006. Ahiaruhe gravel as in photo B, interrupted by a 2 m band of water-laid sand silt and claywhich is conformable at base, channeled by erosion at top, and signifies a hiatus in fluvial aggradation,possibly an interstadial climate interval. D, Ahiaruhe Road, type locality of Ahiaruhe Formation,S27 274094. Ridge Road 2 Loess, Mount Curl Tephra and Penultimate Interglacial paleosol.Downloaded by [Washington State University Libraries ] at 00:53 16 October 2014 478 Journal of The Royal Society of New Zealand, Volume 26, 1996brown gravel with columnar jointed silt (loess?) at middle (Collen & Vella 1984, Fig. 15); 2)gravel quarry, White Rock Road (site 6, Fig. 1) c.10 m of yellow to red brown channel cross-bedded gravel, no internal fine-grained layers, three Last Glaciation loess layers and LastInterglacial paleosol and sand overlying with angular unconformity (Fig. 4B); 3) road cuttingsouth side of Featherston-Martinborough Road at NW flank of Bidwill Ridge, 8 km east ofFeatherston, 2 m yellow brown pebble-cobble gravel above gutter overlain by 2 m greywater-laid sand and mud with weak soil development in turn overlain by exposed 6 m ofyellow to red brown pebble-cobble gravel (Fig. 4C) above which is mainly covered AhiaruheFormation to the top of Bidwill Trig, elevation 112 m, 2 km south, indicating a >70mthickness of the formation. The SE side of Bidwill Trig is an eroded scarp of a no longeractive fault (here named Te Maire Fault after the adjacent Te Maire Stream) down on the SEside and probably eroded back by the Ruamahanga River when it was flowing across its LastInterglacial flood-plain; Francis Line Formation is inferred to overlap the trace of the fault.No terraces of Ahiaruhe age have been found on the floor of the southern WairarapaValley. The top of Ahiaruhe Formation is still distinguishable as the tectonically deformedand deeply dissected aggradation gravel Ahiaruhe Surface on hills forming the east side ofPonatahi Valley east of the present Wairarapa Valley floor, approximately defined bystructure contours (Collen & Vella 1984, Fig. 10). The aeolian deposits, loess, tephra andpaleosol within the formation show that terraces older than the Ahiaruhe Surface existedabove river flood-plain levels within Ahiaruhe time but have been buried under youngerAhiaruhe alluvial deposits because of the tectonic subsidence that prevailed. The stratigraphyis not well enough known to work out a complete sequence of either alluvial aggradation orloess (i.e. stadial) units.The inter-regional correlation of Ahiaruhe Formatiuon depends on the presence above itof the full sequence of Last Glacial and Last Interglacial Stage deposits, and within it ofMount Curl Tephra at two sites, the type locality at Ahiaruhe Road, and 4 km to the west inthe south bank of the Ruamahanga River beside Foreman-Jury Road (Collen & Vella 1984).At the type locality light yellow brown silt below the tephra has vertical joints separatingpolygonal columns characteristic of subaerially deposited loess recognised by C.G.V. in1993 (Fig. 4D). The loess and tephra were deposited subaerially on a terrace tread. Brownclay and sandy clay overlying the tephra is likewise subaerial, being a paleosol modified byburial under thick gravel. The colour and high clay content of the paleosol indicate a largeproportion of aeolian iron-rich andesitic ash in contrast to the iron-poor rhyolitic Mount CurlTephra. The same sequence is seen in the Ahiaruhe Formation at the Ruamahanga Riverbeside the Foreman-Jury Road (Fig. 2) but is more difficult to examine because most of it isunder water. At both sites the paleosol is abruptly overlain by upper Ahiaruhe gravel.INTER-REGIONAL CORRELATIONFig. 2, above, shows correlations of the southern Wairarapa stratigraphic sequence to thereference section at Ridge Road adjacent to the Pahiatua refuse tip, northern Wairarapa(Vella et al. 1988). The Pahiatua section includes eight loess units and two prominentstrongly coloured iron-rich paleosols. The upper paleosol is overlain by three loess units,each related to different aggradation gravel terraces, and with interbedded Aokautere Ashand the andesite rich tephric paleosol corresponding to the Middle Tongariro Tephras ofMilne (1973). The three loesses correlate directly to the Porewan, Ratan and Ohakeanaggradation terrace-loess sequence in southern Wairarapa. The upper paleosol underlyingthose three loesses has been correlated on stratigraphic grounds with the Last InterglacialStage, and correlates to the Francis Line Formation and associated paleosols in southernWairarapa.Three older loess units overlie the lower prominent paleosol at Pahiatua. Of these only thetwo upper ones are related to known aggradation gravel terraces (Vella et al. 1988). Directlybelow this paleosol is Mount Curl Tephra, equally prominent, with two loess layers underlyingit. No loess layers or aggradation gravel terraces in southern Wairarapa can be correlatedDownloaded by [Washington State University Libraries ] at 00:53 16 October 2014 Vucetich Glacial deposits in southern Wairarapa 479with any of the three loesses overlying the paleosol, but correlations can be made to the loessunits underlying the Mount Curl Tephra.The thin-layered sequence of a prominent paleosol above Mount Curl Tephra above loessis known widely in southern North Island inland from the west coast and is the keystratigraphic marker that enables correlation of Ahiaruhe Formation to the Pahiatua referencesection. The loess below the tephra at Ahiaruhe Road (Fig. 4D) and Foreman-Jury Road (Fig.2) correlates to Ridge Road 2 Loess at Pahiatua and Aldworth Loess (Milne 1973) atRangitikei. The underlying gravel and silt of the Ahiaruhe Formation, best but poorlyexposed at Foreman-Jury Road (Fig. 2) must have been a terrace when the Ridge Road 2Loess, tephra and paleosol were deposited, but the gravel would have been a loess sourcewhen aggrading and presumably correlates, at least in part, with Ridge Road 3 Loess andWaituna Loess (Milne 1973).In Wellington district Milne (1980, p. 58, Fig. 28) compared bulk density and watercontent (% dry weight) in continuous cores taken through cover beds at Horokiwi Valley,Haywards Hill and Craigs Flat, showing a sequence of seven loess units best defined in theHorokiwi core. As at Pahiatua, six loess units overlie Mount Curl Tephra (Mangaroa Ash).One loess underlies it. The six loess units with intervening paleosols above the tephra matchthose at Pahiatua. The one loess below the Mount Curl Tephra and the Craigs Flat Terracegravel below the loess correlate with the lower Ahiaruhe gravel and the Ridge Road 2(Aldworth) Loess.Fossil pollen associated with Mount Curl Tephra at Ohariu Valley, near Wellington(Mildenhall et al. 1977) indicate varying climate. Loess close below the tephra yielded acold-climate grass-dominated assemblage. The tephra yielded no pollen. A tephra-rich caporiginally described as reworked tephra, but in fact a paleosol yielded poorly preservedpollen with a clearly warm climate aspect, including common fern spores and rimu pollen,some beech (fusca group) and sparse grass. Loess above the paleosol yielded a cold-climateassemblage but less dominated by grass than that from the loess below the tephra. The unitsrepresented are correlated, in ascending stratigraphic order, with Ridge Road 2 Loess, MountCurl Tephra, Penultimate Interglacial Paleosol and Ridge Road 1 Loess at Pahiatua (Fig. 2).Geological age correlations of the Last Glacial and Last Interglacial deposits are notdisputed. Those of the older deposits are, because recent estimates of the numerical age ofMount Curl Tephra (Kohn et al. 1992; Alloway et al. 1993) suggest that the paleosoloverlying the tephra represents the Antepenultimate Interglacial Stage, not the PenultimateInterglacial Stage as formerly proposed (e.g. Vella et al. 1988). The difficulty arising fromthat correlation is that it leaves no prominent paleosol to represent the Penultimate InterglacialStage, and only three fluvial aggradation related quartz-feldspar loess units separated byweak paleosols to represent the Penultimate Glacial, Penultimate Interglacial andAntepenultimate Glacial stages. Such a correlation is highly improbable.ACKNOWLEDGEMENTSSally Rowe, Research School of Earth Sciences, Victoria University of Wellington,drafted two of the figures. John Nalder, Ned Pattle and other members of Mana U3 Aaided much of the laborious work of augering in Wairarapa.REFERENCESAlloway, B. V.; Pillans, B. J.; Sandhu, A. S.; Westgate, J. A. 1993: Revision of the marine chronologyof the Wanganui Basin, New Zealand, based on the isothermal plateau fission-track dating of tephrahorizons. Sedimentary geology 82: 299-310.Arnott, M. J. 1989: Geology of the Waihora Stream area, western Aorangi Range, Wairarapa. B.Sc.Honours Thesis, Geology Dept. Victoria University of Wellington.Berger, B. W.; Pillans, B. J.; Palmer, A. S. 1992: Dating loess up to 800 ka by thermoluminescence.Geology 20: 403-406.Downloaded by [Washington State University Libraries ] at 00:53 16 October 2014 480 Journal of The Royal Society of New Zealand, Volume 26, 1996Cape, C. D.; Lamb, S. H.; Vella, P.; Wells, P. E.; Woodward, D. J. 1990: Geological structure ofWairarapa Valley, New Zealand from seismic reflection profiling. Journal of the Royal Society ofNew Zealand 20: 85-105.Collen, J. D.; Vella, P. 1984: Hautotara, Te Muna and Ahiaruhe Formations, middle to late Pleistocene,Wairarapa, New Zealand. Journal of the Royal Society of New Zealand 14: 297-317.Cowie, J. D., 1964: Loess in the Manawatu district, New Zealand. New Zealand journal of geology andgeophysics 16: 587-609.Ghani, M. A. 1978: Late Cenozoic crustal movements in the southern North Island, New Zealand. NewZealand journal of geology and geophysics 21: 117-125.Kennett, J. P. 1982: Marine geology. Prentice-Hall, 813pp.Kohn, B. P.; Pillans B.; McGlone M. S. 1992: Zircon fission-track age for middle PleistoceneRangitawa Tephra, New Zealand, stratigraphic and paleoclimatic significance. Paleogeography,paleoclimatology, paleoecology 95: 73-94.Lamb, S. H.; Vella, P. 1987: The last million years of deformation in part of the New Zealand plateboundary zone. Journal of structural geology 9: 877-891.Mildenhall, D. C.; Williams, D. N.; Seward, D. 1977: Ohariu tephra and associated pollen bearingsediments near Wellington, New Zealand. New Zealand journal of geology and geophysics 20: 157-164.Milne, J. D. 1973: Upper Quaternary geology of the Rangitikei drainage basin. PhD Thesis, VictoriaUniversity of Wellington.Milne, J. D. 1980: Surficial deposits. In W. B. Healy, coordinator, Pauatahanui Inlet an environmentalstudy. New Zealand Department of Scientific and Industrial Research, DSIR Information Series141: 57-65.Milne, J. D. 1988: Loess facies in New Zealand (abstract). In D. N. Eden and R. J. Furkert, editors:Loess, its distribution, geology and soils, p. 141. Balkema, Rotterdam.Palmer, A. S. 1982: The stratigraphy and selected properties of loess in Wairarapa, New Zealand. PhDThesis, Victoria University of Wellington.Palmer, A. S.; Vucetich, C. G. 1989: Last Glacial loess and early Last Glacial vegetation history ofWairarapa Valley, New Zealand. New Zealand Journal of Geology and Geophysics 32: 499-513.Te Punga, M. T. 1952: reprinted and dated 1953 after nearly all the first printing was destroyed by fire.The geology of Rangitikei Valley. New Zealand Geological Survey Memoir 8. 46pp. 3 maps inpocket.Te Punga, M. T. 1959: Rangitawa Pumice. In C. A. Fleming editor, Lexique StratigraphiqueInternationale, Fasc. 6 vol. 4: p. 347.Vella, P. 1963: Upper Pleistocene succession in the inland part of Wairarapa Valley, New Zealand.Transactions of the Royal Society of New Zealand, Geology 2: 63-78.Vella, P.; Kaewyana, W.; Vucetich, C. G. 1988: Quaternary terraces and their cover-beds, north-western Wairarapa, New Zealand, and provisional correlations with oxygen isotope stages. Journalof the Royal Society of New Zealand 18: 309-324.Warnes, P. N. 1989: The Quaternaiy geology of the area east of Carterton, Wairarapa, New Zealand.MSc Thesis, Victoria University of Wellington.Warnes, P. N. 1992: Last Interglacial and Last Glacial Stage terraces on the eastern side of WairarapaValley between Waiohine and Waingawa Rivers. Journal of the Royal Society of New Zealand 22:217-228.Wilson, C. J. N.; Switsur, V. R.; Ward, A. P. 1988: A new C-14 age for the Oruanui (Wairakei)eruption, New Zealand. Geology 125: 297-300.Received 22 December 1995; accepted 15 April 1996Downloaded by [Washington State University Libraries ] at 00:53 16 October 2014 Vucetich Glacial deposits in southern Wairarapa 481APPENDIXTable 1 Previously reported sites with last interglacial depositsBy Warnes(1989, 1992):Al: Francis Line Road 200 m SE from junction of Dorset Road, NZMS 260 Sheet S26 (Carterton)metric grid Ref. 263176, exposure in road cutting.A2: Bristol Road 300 m N of junction of Francis Line Road, S26 283163, augered.B: Rutland Road, east edge of Carterton, former Carterton Brickworks quarry, S26 230163, exposure inquarry wall, base now covered by water.C: Carter's Line Road, 20 m Sof house, S26 267151, augered.D: L. Cairns' property 400 m E. of Woodlands Road, S26 273141, augered.E: Udys Road, base of hill 300 m E of junction with Marshall Road, S26 218136, augered.F: Crest of ridge 5 m S of end of Udys Road, S26 220234, augered. G: 300 m E of Woodlands Road,1.2 km SW of site D, S26 261134, augered.By Palmer and Vucetich (1989):H: Bidwill Ridge E side of Martinborough-Greytown Road, S27 164043, continuous core taken withtruck-mounted drill.I: E side of Martinborough-Lake Ferry Road, 500 m S of Otaraia Trig, S27 054885, continuous core asabove.By Arnott(1989):J: Waihora Stream lkm S of Martinborough-Lake Ferry Road 8.4 km NE of Pirinoa township, S27 015862,fossil record f606, exposure.Table 2 New sites1: Ponatahi Road 300 m S of junction with Millars Road, T27 257082, partly exposed in cutting, partlyaugered.2: S side of woolshed 400 m NE of Millars Road, T27 276080, augered.3: E side of Ponatahi Road 800 m N of junction of Foreman-Jury Road, T27 247077, partly exposed incutting, partly augered.4: E side of Martinborough-Greytown Road 20 m N of entrance to Waitiro homestead S27 158029,partly exposed, partly augered.5: SE side of White Rock Road 1.7 km from junction with Martinborough-Lake Ferry Road, S27 153931,augered.6: Top of gravel quarry, W side of White Rock Road 400 m NE of Harris Trig, S27 165916, exposure.7: W side of White Rock Road 600 m S of junction of Te Muna Road, S27 163883, augered.8: 40 m E of White Rock Road nearly opposite site 7, S27 164883, augered.9: N side of Featherston-Martinborough Road 7.6 km E of Featherston, S27 106014 poorly exposedcover beds above Ahiaruhe gravel.10: W side of Featherston-Martinborough Road at junction of Greytown Road, S27 143002, exposedcover beds, sequence obscured by slumping, overlying Ahiaruhe Formation.11: 100 m N of Pahautea Road, 450 m E of Kahutara Road, S27 082960, abandoned quarry, cover bedsoverlying Aharuhe Formation deformed tectonically with decimetre to metre displacement normalfaults down to SE.Table 3 Examples of sections showing loess facies and thickness variationsSite 4, Waitiro (Table 2):0 to 0.2 m A & B horizonsto 1.6 m OHAKEA LOESS: moderately weathered clay loam mottled 7.5YR 8/1, light grey and 7.5YR6/4, dull brown; Aokautere Ash at 1.2 m.to 2.8 m RATA LOESS: friable clay loam 2.5YR 4/7, reddish brown with 10YR 6/8, bright yellow-brown mottles, small degraded plant remains; passing down to plastic clay 5GY 8/1, light grey (0.5 m)Downloaded by [Washington State University Libraries ] at 00:53 16 October 2014 482 Journal of The Royal Society of New Zealand, Volume 26, 1996with small black degraded plant remains, to 5.0m POREWA LOESS: plastic clay mottled 7.5YR 6/8,orange and 7.5YR 7/3, dull orange, grading down to 7.5YR 8/2, light grey and 7.5YR 7/3, dull orangeclay, rare degraded plant remains (0.8 m); friable clay loam 5YR 6/45/8, bright brown grading downto plastic clay loam 7.5YR 8/2-5YR 6/4, grey brown to dull orange (0.8 m); firm clay 5BG 5/1, bluegrey, with 7.5YR 4/2 greyish olive mottles grading down: to 5.4 m LAST INTERGLACIAL PALEOSOL:friable crumbly clay loam, 7.5YR 4/4, dark brown, iron-rich, unconformably? overlying subroundedgravel.Site 5, White Rock Road (Table 2): 0 to ca 0.2 m modified roadside, to 1.7 m OHAKEA LOESS: firmclay loam 10YR 7/3, dull yellow orange with 10YR 5/8, yellow brown mottles, becoming firmer anddarker, 10YR 6/4 towards base. Aokautere Ash not detected, to 3.1 m RATA LOESS: paleosol (0.1m)sandy silt loam 2.5YR 6/46/6 dull to bright yellow brown; grades down to friable sandy loam 5YR 5/8bright red brown; grades down to friable sandy loam 7.5YR 6/6 orange, to 5.1m POREWA LOESS:Lake Ferry (Middle Tongariro) Tephric Paleosol (0.2 m) sandy silt loam 7.5YR 5/8 bright brown;grades down to clay loam 2.5YR 8/1 light grey with 7.5YR 5/6 bright brown mottles; becomes moreclayey towards base with 5YR 5/6 bright brown mottles, to 5.6 m LAST INTERGLACIAL DEPOSITS:paleosol (0.2 m) friable silty clay 10YR 5/6 yellow brown; grades down to weathered clay withrhizomorphs 7.5YR 7/1 light brown grey. (Water bore 100 m NW showed 30 m of grey mudstoneoverlying gravel aquifer).Site D, Woodlands Road (table 1): 0 to 0.3 m A and B horizons, to 0.8 m OHAKEA LOESS: plasticsilty clay loam 10YR 8/1 light grey with minor 10YR 6/8 yellow brown mottles; Aokautere Ashdetected microscopically 0.3-0.4 m. to 1.4 m RATA LOESS: plastic friable to fairly firm silty clayloam 10YR 6/8 bright yellow brown, lcm black concretions in lower half.to 2.0 m POREWA LOESS: Lake Ferry (Middle Tongariro) Tephric Paleosol (0.2 m) silty clay loam10YR 5/6 bright brown, many black concretions; grades down to sticky plastic clay to silty clay loammottled 7.5YR 5/1 brownish grey and 7.5YR 5/6 bright brown.to 2.3 m LAST INTERGLACIAL DEPOSITS: paleosol sticky plastic clay to silty clay 7.5YR 5/6bright brown; grading down to light brown grey fine-grained alluvium (Francis Line Formation).Downloaded by [Washington State University Libraries ] at 00:53 16 October 2014