A model of tephra dispersal from an early Palaeogene shallow submarine Surtseyan-style eruption (s), the Red Bluff Tuff Formation, Chatham Island, New Zealand

Sedimentary Geology 300 (2014) 86–102

Contents lists available at ScienceDirect

Sedimentary Geology

j ourna l homepage: www.e lsev ie r .com/ locate /sedgeo

Amodel of tephra dispersal from an early Palaeogene shallow submarineSurtseyan-style eruption(s), the Red Bluff Tuff Formation, ChathamIsland, New Zealand

Leonor Sorrentino ⁎, Jeffrey D. Stilwell, Chris MaysBuilding 28, Wellington Rd., Clayton Campus, School of Geosciences, Monash University, Victoria 3800, Australia

⁎ Corresponding author. Tel.: +61 3 9905 1642; fax: +E-mail addresses: [email protected], leon

(L. Sorrentino).

0037-0738/$ – see front matter © 2013 Elsevier B.V. All rihttp://dx.doi.org/10.1016/j.sedgeo.2013.12.001

a b s t r a c t

Article history:Received 6 July 2013Received in revised form 3 December 2013Accepted 5 December 2013Available online 18 December 2013

Editor: B. Jones

Keywords:PalaeogeneTephraVolcanic sedimentologyFacies analysisSurtseyan

The Red Bluff Tuff Formation, an early Palaeogene volcano-sedimentary shallow marine succession from theChatham Islands (NewZealand), provides a unique framework, in eastern ‘Zealandia’, to explore tephra dispersalprocesses associated with ancient small phreatomagmatic explosions (i.e. Surtseyan-style eruptions). Detailedsedimentological mapping, logging and sampling integrated with the results of extensive laboratory analyses(i.e. grain-size, componentry and applied palaeontological methods) elucidated the complex mechanisms oftransport and deposition of nine identified resedimented fossiliferous volcaniclastic facies. These facies recordthe subaqueous reworking and deposition of tephra from the erosion and degradation of a proximal, entirely sub-merged ancient Surtseyan volcanic edifice (Cone II). South of this volcanic cone, the lowermost distal facies pro-vides significant evidence of deposition as water-supported volcanic- or storm-driven mass flows (e.g. turbiditycurrents andmud/debris flows) of volcaniclastic and bioclastic debris,whereas the uppermost distal facies exhib-it features of tractional sedimentary processes caused by shallow subaqueous currents. Further north, within theproximity of the volcanic edifice, the uppermost facies are represented by an abundant, diverse, large, and wellpreserved in situ fauna of shallow marine sessile invertebrates (e.g. corals and sponges) that reflect theprotracted biotic stabiliszation and rebound following pulsed volcanic events. Over a period of time, these stableandwave-eroded volcanic platformswere inhabited by a flourishing and diversifyingmarine community of ben-thic and sessile pioneers (corals, bryozoans, molluscs, brachiopods, barnacles, sponges, foraminifera, etc.). Thissuccession exhibits a vertical progression of sedimentary structures (i.e. density, cohesive and mass flows,and cross-bedding) and our interpretations indicate a shallowing upwards succession. This study reportsfor the first time mechanisms of degradation of a Surtseyan volcano on Chatham Islands and contributes to abetter understanding of complex ancient volcano-sedimentary subaqueous terrains. This model of deposition(i.e. onlapping/overlapping features onto the remains of volcanic edifice(s), a vertical transition of structuresfrom deeper- to shallower-marine environments, disaster faunas and subsequent preferential colonisation ofdiverse biota, including large in situ sessile invertebrates, on the summit), characterises an extraordinary exam-ple to be applied to other ancient subaqueous volcanic environments.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

Discriminating primary volcanic deposits from resedimented faciesis a challenging problem faced by researchers of submarine volcano-sedimentary sequences (Bennett, 1972; McPhie et al., 1993), such asthe Red Bluff Tuff Formation (herein RBT), Chatham Islands, NewZealand. Moreover, facies of ancient volcanic successions exemplifythe spectrum of processes that occur during volcanic events, such asclast formation, transport and deposition, as well as post-volcanicsedimentary processes (Brown et al., 1994; Carey et al., 1996; Cantelliet al., 2008).

61 3 9905 [email protected]

ghts reserved.

In general, volcaniclastic submarine aprons are often the onlyremaining evidence of submarine volcanic activity (Soh et al., 1989),especially when the volcanic edifice has been eroded following thecessation of volcanism. Therefore, enhanced understanding of thesevolcaniclastic and resedimented epiclastic facies is important in termsof reconstructing marine palaeoenvironments that were affected byvolcanic activity (e.g. Allen et al., 2007).

In view of the abovementioned relevance to improving our compre-hension of marine volcanic palaeoenvironments, ample studies havebeen undertaken in the last 25 years on volcaniclastic sedimentationof arc-related basins in México (White and Busby-Spera, 1987), USA(Busby-Spera, 1988), and New Zealand (Houghton and Landis, 1989);welded Ordovician tuffs in the United Kingdom (Fritz and Howells,1991); and volcaniclastic aprons in New Zealand (Allen et al., 2007)and Montserrat (Trofimovs et al., 2006, 2008), amongst others.

87L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

However, assessing the complexity of volcaniclastic deposition, due tovariations in eruption styles, bathymetry of marine basins and changingvolcanic slope angles, is a major challenge aswe expand our knowledgein depositional environments and processes that could have formedthese deposits. Good preservation and access of coastal exposures, andthe lateral and vertical variations of facies of the RBT succession, providea unique and exciting opportunity to reconstruct the two-dimensionalarchitecture, structure, transport, and deposition processes during shal-lowmarine clastic sedimentation. The results of this investigation allowto a better understanding of complex ancient volcano-sedimentarysubaqueous terrains. In this paper, we describe in detail the syn- andpost-volcanic facies from the submarine RBT (upper Palaeocene–lowerEocene) relating to pulsed volcanic events along the southwestcoast of Chatham Island, and this model of deposition represents acomplete, interesting and novel example to be applied to other ancientsubaqueous volcanic environments. More importantly, it reportsfor first time bioclastic and volcaniclastic sedimentation associatedwith monogenetic Surtseyan volcanism from the Chatham Islands,New Zealand.

2. Geological setting

The Chatham Islands are the only emergent part of the ChathamRise, a continental submarine plateau extending east from South Island,New Zealand (Fig. 1). The Chatham Rise, along with the rest of the NewZealand subcontinent (New Zealand, Campbell Plateau, Challenger Pla-teau, Lord Howe Rise, Norfolk Ridge, andNewCaledonia)was separatedfrom the continental margin of Gondwana by rifting and extension inthe mid-to Late Cretaceous (Larter et al., 2002; Eagles et al., 2004;

Fig. 1. Location of the Chatham Islands as an emergent segmen

Laird and Bradshaw, 2004; Stilwell and Consoli, 2012; Mays andStilwell, 2013). Specifically, Sutherland (1999) and Eagles et al. (2004)place the opening of the Tasman Sea and separation of New Zealandsubcontinent from West Antarctica (Marie Byrd Land) at 85–80 Ma;however, recent and more refined studies estimate this timing to83 Ma (Campbell and Hutching, 2007; Tulloch et al., 2009). The Chat-ham Rise region (including the Chatham Islands) drifted northwardfrom ~70 to 80°S in the Early Cretaceous to 54°S at the K–Pg boundary(Sutherland, 1995; Stilwell et al., 2006; Timm et al., 2010; Stilwell andConsoli, 2012). The lower Cenozoic units are thinmarine successions as-sociatedwith sporadic volcanism, deposited during regional subsidenceof the ChathamRise and contemporaneous fragmentation of Gondwana(Grindley et al., 1977; Campbell et al., 1993). Marine conditionsprevailed with accumulation of limestone, and a period of non-deposition and erosion marked the end of the early Cenozoic. This un-conformity can be traced along the Chatham region (Austin et al.,1973; Herzer and Wood, 1988; Wood et al., 1989). During the Paleo-gene, the Chatham Islands continued migrating north from approx.latitude 50°S (Molnar, 1975; Grindley et al., 1977) to its present positionat 44°S. The stratigraphy of the Chatham Islands includes a well-preserved record of Late Cretaceous high-latitude Gondwana flora,fauna and environments (Stilwell et al., 2006; Pole and Philippe, 2010;Consoli and Stilwell, 2011;Mays and Stilwell, 2012), as well as an isolat-ed post-Gondwana break-up volcanic oceanic archipelago in the South-west Pacific Ocean during the Paleogene ( Hay et al., 1970; Campbellet al., 1988; Stilwell, 1997; Sorrentino et al., 2011).

The Red Bluff Tuff Formation comprises a succession of upper Paleo-cene to lower Eocene volcanic and volcaniclastic deposits. It was initiallydescribed byHay et al. (1970), and redefined and included as part of the

t of the Chatham Rise, east of South Island, New Zealand.

88 L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

Kekerione Group by Campbell et al. (1988). The RBT includes the re-mains of a cluster of Palaeogene Surtseyan cones, and the completestratigraphic section of the cones consists of two sub-sections: 1) alower sub-section representing volcanic aggradational processes thatconstructed tuff cones in a short period of time (approximately years,based on modern analogues), and is composed of a bedded interval ofexplosively fragmented, vesicular, glassy, basaltic pyroclasts (ash tolapilli sizes) as well as non-explosive emplaced feeder-dykes, pillow-lavas and pillow-sills; and 2) an upper sub-section (i.e. the focus ofthis paper) representing denudation of these cones by shallow marinecurrents or gravity-flows, reflecting the instability of the tephra-pileforming the cones, and a later marine faunal colonisation stagecharacterised by an abundant and diverse fossil assemblage comprisingcorals, barnacles, echinoids, brachiopods,molluscs, bryozoans, and fora-minifera (Sorrentino et al., 2011).

3. Methods

Two field seasons during January and November 2008were conduct-ed on the Chatham Islands, whereby detailed mapping of the RBT wasundertaken and representative rock samples were collected. The fossilif-erous volcaniclastic facies described in this paper are defined on the basisof field characteristics including composition, texture and depositionalstructures combined with the results of laboratory analysis of compo-nents and textures, following McPhie et al. (1993) and Cas et al. (2009).

Deposit grain-sizes were estimated through a semi-quantitativeanalysis following the methods outlined in Sorrentino et al. (2011).The results from initial field observations were compared to scans oftwo representative thin-sections per facies for average and statisticalvalidity. A digital grid of up to 15 vertical and horizontal divisions(up to 225measurements in total) was placed over the images. Each in-dividual clast intersected by the grid was counted and its size estimatedby comparison with digitised circles of the following sizes: 0.125 mm,0.25 mm, 0.5 mm, 1 mm, 2 mm, 4 mm, 8 mm, and 16 mm diameter.Finally, the results were plotted using the program Gradistat by Blottand Pye (2001), with the purpose of assessing various textural parame-ters, e.g. sorting, skewness and grain-size distribution. The grain-sizecategories are as follows: mud is b0.0625 mm (comprising both siltand clay-sized particles); sand is 0.0625 to 2 mm; grit is 2 to 4 mmand pebble is 4 to 64 mm(unless otherwise specifiedwhere the catego-ries are different). Percentage values have uncertainties of up to ±1%.

Over 30 thin-sections, representing RBT volcaniclastic fossiliferousfacies, were studied. Quantitative percentage abundances of the compo-nents of each lithofacies incorporating fragments of fossils, volcanicclasts, free crystals, and matrix (in part slightly altered to cement),were determined employing thin-section point counting on a petro-graphic microscope. The point counting included 400 counts per sliderepeated three times on each thin-section to minimise errors andthese were presented as main results. Also, all vesicle diameters wereestimated from thin sections, and the diameters of these vesicles werecorrected for by the shareware software package Image J (Schneideret al., 2012).

Some localities (e.g. Pukekio) are labelledwith names proposed pre-viously by Campbell et al. (1993). Additionally, we have labelled new lo-calities using informal names (e.g. “Sheep Terrace” and “Small Spot”) forease of reference throughout the present study.

Lastly, a geological hammer or field notebook are included as scale inmost figures illustrated below. The length of the hammer and the fieldnotebook are approximately 30 cm and 20 cm, respectively.

4. Results

This section contains a detailed description of nine facies document-ed from the southernmost outcrops of the RBT Formation (see encircledarea in Fig. 1). In addition, the lateral distribution of these facies,with re-spect to the source (Surtseyan volcanic Cone II; PointWeeding location,

43º56′39.2″S; 176º34′20.2″W) is identified, which will aid to depict therange of dispersal processes (Fig. 2).

4.1. Facies analyses

The complete stratigraphy of RBT is divided into two facies associa-tions (Sorrentino et al., 2011). Firstly, a primary volcanic facies associa-tion with abundant moderate- to poorly-vesicular, irregular andcuspate-shaped (glass shards that retain the original shape of thebubble-wall), sand-sized (coarse-ash) volcanic clasts fragmented ex-plosively and deposited mainly by fall-out and down-slope dilutegravity-flows in shallow marine environments; these were succeededby the intrusion and extrusion of pillow-sills and pillow-lavas, respec-tively. These facies developed several small volcanic edifices, locatedclose to each other, where the volcanic activity persisted repeatedlyand without major interruptions or stratigraphic breaks duringmonths or even several years (by comparison with modern analoguesas Surtsey, Iceland). Secondly, a fossiliferous volcaniclastic facies associ-ation with abundant sub-angular volcanic clasts, interpreted to bederived from early erosion and redeposition due to shallow marinecurrents, instability on slopes caused by further volcanic activity and/or seismic tremors, and copious fragments of marine fossils, whichpreserve evidence of a later ample and diverse faunal colonisation ofthe eroded and stable volcanic platforms.

The primary volcanic facies association has been described compre-hensively in Sorrentino et al. (2011). As a result, the following descrip-tion of facies refers to the syn- and post-depositional volcaniclasticfacies association. It is important to note that these facies are describedin stratigraphic order, starting from the lowermost facies, cropping outand designated as CH1 and ending with the uppermost facies CH26.Following each facies description a brief interpretation is included foreffortless reading. An exhaustive discussion of these interpretations isexplained in Section 5.

4.1.1. Calcareous and muddy laminated ash-rich mudstone (Facies CH1)This is the lowermost outcropping facies, and the basal contact of

this facies with older deposits within the RBT is not exposed (Fig. 3). Fa-cies CH1 is overlain by Facies CH2 at Pukekio (43°57′52.8″S; 176°35′03.4″W), located south of the Surtseyan volcanic Cone II (Sorrentinoet al., 2011). The contact between these two facies is erosive and mark-edly irregular, reflecting the rapid deposition of overlying sediments(Facies CH2) on top of unconsolidated Facies CH1 (Fig. 4b). There is nopreserved paleosol, karstification or other evidence of subaerial expo-sure. The erosion of the underlying CH1 is inferred to have occurredconcurrently with deposition of CH2.

Each layer is planar and laterally continuous for a minimum lengthof 3 m in outcrop. These layers appear to be mantling or coveringolder structurally irregular or uneven rock formations, due to the undu-lating sedimentary structure of CH1. The preserved thickness of this fa-cies is approximately 1 m as represented at Pukekio, the only accessibleoutcrop thus recorded.

The main depositional structures present are very thin diffuse bed-ding (1–3 cm) and abundant diffuse lamination (b1 cm). The textureof this facies is characterised by abundant mud-sized particles 55.7%,with less abundant (43.3%) sand-sized particles, and minor (0.5%) grit-sized grains (Table 1). The fabric is matrix-supported (wackestone).The sorting of this facies is poor (Ф = 1.661), reflecting a transportmechanism or duration that was not conducive to develop high levelsof size-sorting. The roundness of the clasts varies from sub-angular tominor sub-rounded and extremely limited fossils (i.e. indistinguishableforaminifera and small fragments ofmolluscs/barnacles?) are preserved.

Facies CH1 comprises fourmain components: 1) relicts ofmud-sizedclastic matrix, in part altered to crystalline carbonate cement (calcite;52%); 2) abundant volcanic clasts (38%); 3) minor fragments of fossils(5%); and 4) rare free crystals (5%; Table 1). The matrix is mud-sizedand has largely been replaced by carbonate cement. That is, mud-sized

Fig. 2. Topographic and geologic map of the study area, showing remnants of the volcanic edifice (dark grey), and distribution and contacts of RBT resedimented volcaniclastic and fos-siliferous facies (light grey). The lower image is a schematic NE–SW cross-section from “Small Spot” to Pukekio. Cross-section is not to scale. S.L. = Sea Level.

89L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

particles were originally mostlymicrite that has been (at least partially)recrystallised into a cement. This cement also filled pre-existing emptyspaces such as cavities in fossil fragments and vesicles in volcanic clasts,forming calcite amygdales in the latter case. The volcanic clasts are ba-saltic in composition, and are altered to resinous palagonite; such alter-ation pattern remains the same in every facies. These basaltic clasts arevesicular,millimetre-scale and angular, characterisedmainly by cuspateshapes (glass shards that retain the original shape of the bubble-wall).The vesicles in the volcanic clasts have sub-equant, spherical shapeswith diameters between 30 μm and 160 μm. The fossil contentrepresents only 5% of this facies and is characterised by a very lowdiver-sity (‘depauperate) fauna of shallow marine invertebrates. The sizeof these organisms varies from several microns (foraminifera) toseveral millimetres (e.g. shells fragments). The crystal content isminimal, and consists of very scarce, micron-scale sub-rounded quartzand traces of biotite (phlogopite), green hornblende and chlorite. Addi-tionally, it includes abundant millimetre-scale peloidal glauconitegrains. A significant amount of iron oxides (hematite, magnetite, etc.)are present as grain/matrix coatings, which obscure the textural charac-teristics of the rock. This diagenetic pattern is observed in all facies ofthe RBT.

4.1.1.1. Brief interpretation. Facies CH1 is thought to be most likelydeposited by low-concentration density flows.

4.1.2. Calcareous massive ash-rich mudstone (Facies CH2)Facies CH2 overlies Facies CH1 at Pukekio; the contact is irregular

and erosional, and several intrusive injections of volcaniclastic sedimentcorresponding to Facies CH2 can be recognised within the underlyingFacies CH1 (Fig. 4b). The upper contact with Facies CH3 is also irregular.The thickness of this facies is uneven; it is thicker in the erosional de-pressions or concavities (~0.9 m; ‘hummocks’) and thinner on positiverelieves or convexities (~0.7 m; ‘swales’).

The structure of this facies is massive, and the texture is dominatedby mud-sized grains. Most counted grains were mud-sized (70%),with only 27.1% of sand size and minor grit (2.7%; Table 1). This facieshas a matrix supported (wackestone) fabric, and is poorly sorted(Ф = 1.430). The roundness of clasts is quite poor; the edges of theclasts are rough and angular, and it is important to note that somevolcanic clasts provide evidence of fluidal shapes; these very irregular-edged volcanic clasts have been observed only in this facies.

Facies CH2 consists of four main components: 1) relicts of mud-sizeclastic matrix strongly recrystallised (calcite; 62%) 2) volcanic clasts

Fig. 3. General view of resedimented RBT facies at Pukekio (43º57′52.8″S 176º35′03.4″W). a) Note contacts, distribution and structure of each resedimented facies (CH1, CH2, etc).Encircled in the left top side are people for scale. b) Diffuse-bedding and erosive/irregular contact between facies CH1 and CH2. c) Photomicrograph (plane polarised light) of textureand components of Facies CH1, G = peloidal glauconite, V = altered basaltic volcanic clasts (most of yellow-brownish clasts on thin-section), M = mudmatrix (grey dusty), F = fossils(such asmultilocular tests of foraminifera and shells of gastropods), C = crystals (most shinywhite fragments, e.g. quartz and plagioclase). (For interpretation of the references to colourin this figure legend, the reader is referred to the web version of this article.)

90 L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

(15%); 3) fragments of fossils (13%); and 4) rare free crystals (10%;Table 1). As above, the matrix in this facies is in general mud-sized andhas largely been replaced by carbonate cement. The basaltic volcanicclasts are very poorly-vesicular to non-vesicular, millimetre-scale andsub-angular to sub-rounded, characterised mainly by fluidal margins(spatter-like) and minor cuspate shapes. The fossil content comprises13% of this facies, and is characterised by a very low diversity invertebrateassemblages (e.g. barnacles, molluscs and bryozoans) as well as veryscarce relicts of vertebrates (i.e. shark teeth). The size of these organismsvaries from several tens of microns (rare foraminifera) to several

millimetres (e.g. shells of barnacles). The crystal content is low, and con-sists of scarce micron-scale sub-rounded quartz, olivine, pyroxenes, andplagioclase. Additionally, it includes abundant millimetre-scale sub-rounded grains of peloidal glauconite. As for the aforementioned FaciesCH1, this facies exhibits a significant amount of interstitial iron oxides,which partly obscure the textural characteristics of the rock.

4.1.2.1. Brief interpretation. Facies CH2 is interpreted to be a water-supported volcaniclastic mass-flow deposit, emplaced entirelysubaqueously.

Fig. 4. Distribution, contacts and textural characteristics of CH2. a) Uneven thickness of CH2, thicker in the hummocks (~0.9 m) and thinner in the swales (~0.7 m). b) Close-up view oferosional contact, showing volcaniclastic sediment corresponding to Facies CH2within the underlying Facies CH1. c) Photomicrograph (plane polarised light) of typical poorly-sorted tex-ture of this facies, exhibiting an irregular basaltic volcanic clast of 3.5 mm in length, surrounded by smaller fragments of fossils and abundant calcite cement. d) Close-up of a photomi-crograph (plane polarised light); most green and light brownish components are glauconite, and very dark brown grains are basaltic clasts overprinted by iron oxides, the surroundingcalcite cement appears white. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

91L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

4.1.3. Calcareous graded ash-rich sandstone (Facies CH3)Facies CH3 overlies Facies CH2, and is in turn gradationally overlain

by Facies CH5 and truncated by CH4 as expressed in outcrop on thenorthern side of Pukekio. As for the other RBT facies, the preservedthickness of Facies CH3 varies slightly getting thicker towards thesouth-southeast, approximately 1.5 m thick in the southernmost out-crop located at Pukekio. This individual facies can be traced for severalmetres along the lower part of the cliff.

The main sedimentary structural characteristics present are gradingand lamination (Fig. 5). The base of this facies is normal grain-size grad-ed, followed continuously by planar laminated, reverse-graded finesandstone (~50 cm thick).

Texturally, Facies CH3 consists mainly of sand-sized grains (57.2%),with slightly less mud-sized grains (42%) and very minor (0.5%) grit-sized components (~4 mm; Table 1). The fabric is matrix-supported(wackestone). Similar to the previously described bioclastic facies, thesorting of this facies is very poor (Ф = 1.634), reflecting a mechanismand/or duration of transport that was not conducive to develop highlevels of size-sorting. The roundness of the clasts varies from sub-angular to sub-rounded. The fossils, comprising mostly invertebrateand foraminiferal remains, are more abundant towards the top of thisfacies, but are moderately fragmented; it is still possible to recognisemany invertebrate groups due to the good preservation of the individu-al specimens (e.g. some brachiopods and molluscs with the two valvesstill attached and foraminifera with the tests almost intact).

The four main components of Facies CH3 are: 1) relicts of mud-sizeclastic matrix partially recrystallised into carbonate cement (calcite;57%); 2) abundant fragments of fossils (23%); 3) volcanic clasts (13%);

and 4) rare free crystals (7%; Table 1). The textural and compositionalcharacteristics of cement/matrix, volcanic clasts and crystals are compa-rable to the ones previously described for Facies CH1 and CH2; however,Facies CH3 does not contain volcanic clastswith fluidal shapes, as for Fa-cies CH2 instead, volcanic clasts are rather rounded and, to some extent,have smoother edges. On the other hand, the fossil fragments reveal aconsiderable increase of richness (relative abundance) and diversitycompared to the underlying facies CH1 and CH2; corals, ossicles of cri-noids, and ostracods have been recognised from this facies.

4.1.3.1. Brief interpretation. The uninterrupted succession of facies CH3and CH5 (please see description below) is considered as being deposit-ed by turbulent subaqueous, granular, mass-flows.

4.1.4. Massive ash-rich mudstone (Facies CH4)This facies crops out at Pukekio, and has a distinctive channel-like

structure; additionally, both contacts (upper and lower) with otherbioclastic facies are irregular and the lower contact is erosive (Figs. 3,5). Facies CH4 partially overlies both facies CH2 and CH3, and it, inturn, is overlain by Facies CH5. This facies is approximately 2.5 m in itsthickest part.

Facies CH4 is largely massive, but other visible depositional struc-tures resemble those typical of soft sediment deformation such as loadcasts. Additionally, centimetre-scale erosional scours have beenrecognised at the base of this facies (Fig. 5e). Texturally, this facies com-prises abundant mud-size grains (86.5%) and minor extremely fine- tomedium-sand grains (13.4%). The fabric is clast- to matrix-supported(packstone to wackestone), and it is moderate sorted (Ф = 0.976).

92 L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

Similar to the previous described bioclastic facies, CH4 consists of:1) abundant volcanic clasts (70.5%); 2) crystalline carbonate cement(calcite) and/or relicts of mud-size clastic matrix (25.5%); 3) few frag-ments of fossils (3.25%); and 4) very rare free crystals (0.75%; Table 1).The volcanic clasts are essentially altered vesicular volcanic ash(palagonite); these pyroclasts are angular (little-modified to unmodifiedcuspate-shapes). The cement is calcium carbonate and has infilled pre-existing pore spaces such as vesicles in the volcanic ash (as calciteamygdales) and cavities in the scarce fossils. Due to the poor-preservation and limited quantity of fossils fragments, it is difficult to de-termine the fossils' taxonomic groups with accuracy. Nevertheless, thedispersed fragments of fossils found in this facies include very sparsebenthic foraminifera tests and rare fragments of invertebrate shells,

Table 1Structural, textural, and compositional characteristics of the resedimented fossiliferous volcani

probably molluscs or barnacles. With respect to the mineral content,only one free crystal was identified from this facies, is a well-roundedgrain of peloidal glauconite.

4.1.4.1. Brief Interpretation. Facies CH4 is believed to have formed by acohesive flow, where water was an essential interstitial medium; eithera primary volcanic or an epiclastic mud-flow.

4.1.5. Calcareous and muddy cross-laminated ash-rich mudstone(Facies CH5)

This facies has a gradational boundary with Facies CH3, and in part,also conformably overlies the massive ash-rich mudstone Facies CH4.

clastic facies corresponding to the Red Bluff Tuff Formation.

Table 1 (continued)

93L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

Facies CH5has only been recorded fromPukekio,where anapproximatethickness of 3 m is exposed.

The main depositional structure is cross-lamination (b1 cm; Fig. 6);and these comprise cross-lamination sets approximately 1 to 1.5 mthick; however, detailedmeasurements are unfeasible due to hazardousrock exposure. This facies is characterised by mudstone and sandstonewith grain-sizes varying from 51.9% of mud-size to 47.9% of sand-size.The fabric is matrix-supported (wackestone). The sorting of this faciesis poor (Ф = 1.852); as per Facies CH3, this reflects a mechanismand/or duration of transport that was not favourable to produce highlevels of size-sorting. The roundness of the clasts varies from sub-angular to sub-rounded, and the preservation of the fossils is moderate.

The componentry analysis of this facies revealed: 1) abundant(47.9%) cement/matrix; 2) common basaltic volcanic clasts (38.1%); 3)minor well-preserved fossils fragments (13.3%); and 3) very rare crys-tals (0.7%). The textural and compositional characteristics of cement/matrix, volcanic clasts, crystals, and fossil fragments are comparable tothose described for Facies CH3 (we intentionally omit these descriptionsto avoid unnecessary repetition). However, the abundance of fossil frag-ments decreases from 23% in facies CH3 to 13% in Facies CH5, but thefaunal diversity remains relatively constant, with well-preserved valvesof brachiopods, barnacles, molluscs, fragments of sponges, corals,

crinoids and other unidentifiable remains. Furthermore, the volcanicclasts increase considerably from 13% in the underlying facies (FaciesCH3) to 38% in this facies.

4.1.6. Calcareous, cross-bedded, ash-rich sandstone to pyroclast-richlapillistone (Facies CH6)

This facies is located in the upper stratigraphic levels at Pukekiowhere it is readily mappable, but it can also be observed for several me-tres northwards along the coastline towards the Surtseyan volcanicCone II. The lower contact with Facies CH5 is sharp but conformableand the upper contact with Facies CH7 is gradational. The maximumobserved thickness of this facies is approximately 7 m.

Facies CH6 is characterised by cross-bedding in metre-scale cross-bed sets (Fig. 3), these comprise asymmetrical dunes of several tens ofcentimetres in height as recognised at outcrop scale (Fig. 6; indicativeof south-westwards currents), probably with lunate crests. Texturally,this facies is dominated by sand-sized grains, the grain-size analysisrevealing the following abundances: 21% mud-sized particles, 60.9%sand-sized particles and 16.9% grit to very fine pebble-sized particles(2–8 mm; Table 1). This facies is very poorly sorted (Ф = 2.590). Thisis due primarily to the contrast in size between large well-preservedfragments of fossils and very fine fragmented fossils. Furthermore,

Fig. 5. Structural, textural and compositional characteristic of Facies CH3 and CH4. a) Lower contact of Facies CH3with underlying Facies CH2 and normal grading (white triangle) given bythe accumulation of abundant sub-rounded scoria clasts at the base of facies CH3; 15 cm long pencil in the centre of the photograph as scale. b) Photomicrograph (plane polarised light) ofFacies CH3, observe the good preservation of two multilocular foraminifera in the top-left corner, also a fragment of mollusc shell (bivalve; top-centre) and a fragment of echinoid (top-right), the green clast at the bottom is a peoidal glauconite; and the texture is overprinted by the abundant iron oxides (dark brown fragments). c and d) Grading at the top of this faciesshowing a considerable increase of bioclasts of lower density (density of calcite) compare to thedensity of the volcanic clasts (density of basalt). e) Channel-like structure of Facies CH4 anderosive contact with Facies CH3 (indicated by the geologic-hammer); note the presence of slumps at the top of the photo. f) Photomicrograph (plane polarised light) of distinctivemoderately-sorted fabric of Facies CH4; in the centre is a clast of glauconite (peloid); dispersed throughout the thin-section are abundant fine-grained, altered, basaltic volcanic clasts(brown-yellowish small fragments) and surrounded by calcite cement (white). (For interpretation of the references to colour in this figure legend, the reader is referred to theweb versionof this article.)

94 L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

most clasts are sub-rounded, and like the previously mentionedbioclastic facies, alteration has modified original compositions; in par-ticular, basaltic volcanic clasts have undergone palagonitisation andseveral iron oxides have replaced and overprinted pre-existing min-erals. Minor grains of peloidal glauconite were found in this facies.

Componentry analysis reveals that Facies CH6 exhibits a pronouncedincrease in fossil fragments (49%), with both diversity and richness(relative abundance) rising noticeably (in comparison to all previouslydescribed facies). The basaltic volcanic clasts represent 26.5%; theseclasts are vesicular and sub-rounded, the vesicles are sub-spherical

(with minor elongated vesicles) and small (b120 μm in diameter) andmatrix/cement comprises 24.1% of this facies. Facies CH6 it is almostcrystal-free (0.4%). Fossils are well-preserved and their sizes vary be-tween the different groups of fossils; for instance, while tests of forami-nifera have sizes of up to a few millimetres, fragments of bryozoans,corals or valves of molluscs can have up to several tens of millimetresor even more. Most of the micritic matrix has been replaced by calcite,and the basaltic volcanic clasts are similar to those already described(e.g. Facies CH5). While the characteristics of these constituents are asthose mentioned previously, the fossil content is considerably richer

Fig. 6. Structural, textural and compositional characteristic of Facies CH5 and CH6. a) Contact relationship between facies CH5 and Facies CH6. b) Thin cross-beds, a characteristic featureof Facies CH5. c) Close-up view of poorly sorted, muddy fabric of Facies CH5, with a valve of a pectinid bivalve ‘Chlamys (sensu lato) mercuria’. Marwick (1928; 0.9 cm in length) in thecentre. d) Photomicrograph (plane polarised light) of Facies CH5; fine-grained basaltic volcanic clasts (V) appear as yellowish clasts, due to palagonitisation enclosed by a rim of clays;F = fragment of a valve of a centimetre-scalemollusc. e) Set of dune-size asymmetrical cross-bedding, characteristic of facies CH6,field-note for scale. f) Photomicrograph (plane polarisedlight) of Facies CH6; V = vesicular basaltic volcanic clast altered to palagonite, as, most of the volcanic clasts reported in this study; F = abundant well-preserved fossils fragments ofechinoid plates and foraminiferal tests.

95L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

than in any other described bioclastic facies (except for CH26, pleasesee below).

4.1.6.1. Brief Interpretation. Facies CH6 is considered as being formed bydownstream-migrating bedforms produced by unidirectional aqueousflows.

4.1.7. Calcareous bedded ash-rich sandstone (Facies CH7)This facies is located at the uppermost stratigraphic levels at

Pukekio. Similar to Facies CH6, the calcareous bedded ash-rich sand-stone Facies CH7 can be followed northwards along the cliff towardsthe Surtseyan volcanic Cone II; however, the outcrops are inaccessibledue to very hazardous rock faces. Facies CH7 gradationally overliesFacies CH6 at Pukekio, and laterally, the basal contact of this facies can

be traced for tens ofmetres. Facies CH7 is partially covered by unconsol-idated Quaternary(?) sediments.

In general, the thicknesses of Facies CH3, CH5, CH6 and CH7 varynoticeably, with a consistent thickening trend towards the south-southeast; this succession reaches a total maximum thickness of ap-proximately 20 m in the southernmost outcrop located at Pukekio.The succession forms an apron (pinching towards the volcano) and iseffectively thinner close to the volcanic cone (Cone II) and thicker onthe slope breaks.

Themain depositional structure of Facies CH7 is planar bedding, andin most cases layers are thin- (3–10 cm) to medium- (10–30 cm)bedded. The grain-size abundance distribution shows that 38.1% ofgrains are mud-sized, 57.3% are sand-sized and 4.8% are grit-sized(4 mm). As for Facies CH6, the sorting is very poor (Ф = 2.267) and

Fig. 7. Typical features of the uppermost facies located at Pukekio, Facies CH7. a) Gradational contact with the underlying Facies CH6, the entire view covers approximately 2 m of stra-tigraphy. b) Close-up view of a scaphopod related to Dentalium, this taxon is very abundant within this facies. c) and d) Photomicrographs (plane polarised light) of Facies CH7; theyellow-brownish disperse fragments are vesicular basaltic volcanic clasts altered to palagonite; clasts denoted with “F” are well-preserved fossils fragments of echinoids plates andmultilocular foraminifera tests, “C” indicates white carbonate cement.

96 L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

most clasts are sub-rounded (i.e. volcanic, bioclastic and crystals). Addi-tionally, fragments of fossils are well-preserved (Fig. 7).

The componentry analysis disclosed similar percentages of compo-nents as Facies CH6 however, the fossil content decreases to 38.1% inthis facies. The ratio of matrix/cement (30.8%), basaltic volcanic clasts(30.5%) and a dearth of crystals (0.6%) resembles the underlying facies.

4.1.7.1. Brief interpretation. Similarly to CH6, this facies is also interpretedas formed by shallow marine currents, and the deposition could haveoccurred as bedload.

4.1.8. Calcareous normal graded pyroclast-rich lapillistone (Facies CH8)This facies is located further north from Pukekio, at “Sheep Terrace”

(43°57′29.1″S 176°34′31.5″W). The basal contact of this facies is inac-cessible, but it can be traced laterally towards thepreserved volcanic ed-ifice, Cone II, where it is overlapping the primary pyroclastic volcanicdeposits of the RBT (i.e. bedded to cross-bedded coarse-ash andcoarse-tuff Facies “B”; Sorrentino et al., 2011). This angular contact ischaracterised by an unconformable onlap surface between the primaryvolcaniclastic facies and the presently described bioclastic facies (Fig. 8),and corresponds to the deposition of this facies on the slopes break ofthe volcanic cone. As for the Facies CH7, outcrop located further southat Pukekio, Facies CH8 is covered by unconsolidated Quaternary(?) sed-iments. Due to a lack of accessible outcrop contacts, the precise strati-graphic relationship of CH8 with the facies exposed at Pukekio ispresently unknown.

The maximum thickness of this facies is estimated to be approxi-mately 30 m and it becomes relatively thinner towards the slope ofthe volcano. The main depositional structures are very abundant thincross-beds (including minor possible herringbone cross-stratification)

and normal grading. This facies is interbedded by a uniformly thickash-layer (40 cm thick) that can be traced along strike for several me-tres in both north and south directions along the cliff. The origin ofthis layer remains unclear, as it seems too thin to come from an eruptionof Cone II (several tens ofmetres away), or other knownnearby volcaniccones corresponding to the RBT. Moreover, the geochemistry of thislayer displays a marked compositional difference compared to typicalRBT volcanic ash (Sorrentino et al., unpublished data).

The texture of this facies is characterised by a nearly even distribu-tion of mud-sized (38.9%), sand-sized (32%) and gravel-sized particles(28%, grit to fine pebble-size, 2–8 mm; Table 1). The sorting of thisfacies is very poor (Ф = 2.576), and most clasts are sub-angular. Addi-tionally, fragments of fossils are very well-preserved. This facies has aclast- to matrix-supported fabric, and is, therefore, texturally classifiedas packstone to wackestone.

Facies CH8 consists of only three main components: 1) very abun-dant volcanic clasts (60%); 2) abundant fragments of fossils (29%);and 3) crystalline carbonate cement (calcite) and/or relicts of mud-size clastic matrix (11%; Table 1). Similar to the aforementioned de-scribed facies, the matrix in this facies is generally micritic and haslargely been replaced by calcium carbonate cement. The basaltic volca-nic clasts are vesicular, several of which are millimetre-scale, sub-angular, and enclosed by a rim of iron oxides. The fossil content corre-sponds to approximately 30% of this facies and is characterised by ahigh diversity fauna of invertebrates (e.g. corals, sponges, barnacles,molluscs, and bryozoans). A number of fossil assemblages (e.g. coralsand sponges) appear to be in situ. As in Facies CH7 the size of theseorganisms increases considerably in this unit, with some sponges spec-imens reaching centimetre-scale. While facies CH7 and CH8 are similar,the latter can be distinguished by the higher content of basaltic volcanic

Fig. 8. Distribution and features of Facies CH8 located at “Sheep Terrace” and along the cliff exposures towards Cone II. a) Northeast view at the sloping and contact between the primaryvolcanic facies that created the volcanic edifice (Cone II) and the volcaniclastic and bioclastic Facies CH8. b) Distribution of Facies CH8 at “Sheep Terrace”; note the interbedded ash-layerconfirming that explosive volcanic activity persisted in the Chatham Island regionwhilst the resedimented bioclastic facies of RBTwere emplaced. A person for scale at the right side of thephotograph. c) Thin cross-bedding; inside the black circle is a 15 cm long pencil for scale. d) Photomicrograph (plane polarised light) of Facies CH8; a fragment of bryozoan; andsurrounded bywhite is the crystalline cement (sparry calcite); vesicular basaltic clast (left) has elongated vesicles, and has been completely replacedby palagonite, (note large size of fossilfragments and volcanic clasts compared to lower facies).

97L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

clasts, coarser grain-size and presence of abundant thin cross-beds;accordingly, they were most likely emplaced by different depositionalmechanisms.

4.1.8.1. Brief Interpretation. Facies CH8 is characterised by tractionalsedimentary structures and they are considered to be caused by shallowsubaqueous currents.

4.1.9. Bedded calcareous pyroclast-rich lapillistone (Facies CH26)This is the uppermost recognised volcaniclastic and fossiliferous fa-

cies, and it directly overlies the summit of the remains of the volcanicedifice (Cone II) at “Small Spot” (176°34′20″S 43°56′29″W). FaciesCH26 is stratigraphically and topographically higher than all of theaforementioned facies (Fig. 9). The basal contact with the primary vol-canic Facies “B” (Sorrentino et al., 2011) is gradational and there is noknown outcropping facies above Facies CH26 at present.

The maximum preserved thickness of this facies is approximately4.5 m, and it is characterised by medium bedded (3–10 cm) layers ofvolcaniclastic and abundant bioclastic fragments. Texturally, this faciesconsists almost entirely of sand (52.3%) and gravel- to pebble-sized(2–8 mm; 36%) clasts, with a minor mud component (11%; seeTable 1). The sorting of this facies is very poor (Ф = 2.246) and thefabric is clast-supported (packstone).

Similar to Facies CH8, the present facies comprises three maincomponents: volcanic clasts, fragments of fossils and matrix/cement;however, the fossil content is significantly higher (67%), and the per-centage of volcanic clasts (30%) and matrix (3%) decrease considerablyin this facies. The general composition of this facies is similar to FaciesCH7 and CH8; nonetheless, the diversity and richness of fossils rise

dramatically in Facies CH26.Moreover, the size of these fossils is greaterthan in any of the previous mentioned facies (e.g. sponges several tensof centimetres long) and the preservation of these marine organismsis considerably better. Most fossil assemblages appear in situ.

As previously mentioned alteration has affected all facies; calcite,iron oxides and palagonite are amongst the most evident diageneticproducts. Due to different degrees of alteration, facies are more or lessconsolidated, causing the recognition of individual textures and compo-nents to be rather exigent.

4.1.9.1. Brief interpretation. By comparison with the previously men-tioned facies, Facies CH26 is considered to be formed by shallowermarine currents and aggradational sedimentary processes.

5. Depositional processes and tephra dispersal implications

Herewith is a detailed discussion on the sedimentary processes thattransported pyroclastic and bioclastic debris to the slopes of theSurtseyan volcanic Cone II and more distal areas. An interpretation ofthe sequential relationship between volcanic and bioclastic, and withinbioclastic products, is presented. Additionally, each previously de-scribed facies (see Section 4) is interpreted as being deposited by oneor multiple mechanisms, as inferred from the robust RBT data collected.

Comparable with its modern analogue, the submarine volcanic ventSurtla ( Thorarinsson et al., 1964; Kokelaar and Durant, 1983), the depo-sition of the majority of the RBT resedimented facies (e.g. CH1, CH2 andCH3) most likely occurred soon after (or even during) the volcanic pilewas formed, this pile corresponds to the remains of Cone II. This is sup-ported by the lack of major sedimentologic breaks between primary

Fig. 9. Uppermost facies corresponding to remobilised deposits of RBT (Facies CH26), located at “Small Spot”. a) and b) North view of the gradational contact between volcanic deposits(Facies “A” and “B”), that correspond to the remains of the Cone II, and the bioclastic Facies CH26 in the summit of the cone. c) Detail of themediumwell-bedded highly fossiliferous layers.d) Excellently-preserved fragment of poriferan, this facies has a particular abundance of corals and sponges. e) and f) Photomicrographs (plane polarised light and cross-nicols, respective-ly) of Facies CH26, large bryozoan and mollusc shell-fragments; the surrounding cement is sparry calcite (anomalous colour of interference in cross-nicols). Note the increasing size andpreservation quality of fossils fragments, compared to lower bioclastic facies.

98 L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

volcanic facies and resedimented fossiliferous volcaniclastic facies. Par-ticularly, the contacts between Facies “B” and Facies CH8, on the slopes(an onlapping relationship), and Facies CH26, on the summit, of the re-mains of the volcano (Cone II), are irregular and slightly erosive or gra-dational. Additionally, some volcanic clasts recognised in Facies CH2present fluidal margins; this indicates that these clasts were emplacedwhile still in a molten (plastic) state, inconsistent with erosion of thevolcanic edifice and deposition after a prolonged period of time (i.e.millennium-scale duration). Furthermore, delicate and compositionallyunstable components such as glass shards and scorias are typically ab-sent in epiclastic rocks because they have been destroyed byweatheringand transportation (Fisher and Schmincke, 1984); however, these clasts

are abundant in the RBT resedimented facies and support a pyroclasticorigin from a nearby source.

In the same way, all resedimented facies (CH1 to CH26) were mostlikely emplaced continuously one after another as no major unconfor-mities or sedimentologic breaks corresponding to substantial timegaps are found within each bioclastic facies (Fig. 10). Moreover, thepresence of glauconite in a number of facies (Facies CH1–CH7) supporta geologically short to moderate interval, because the initial glauconiticsmectite requires at least 103–104 years to form (Odin and Matter,1981).

The diffuse thin-bedding and lamination, lateral uniformity of thethickness of the layers and the aspect of mantling the pre-existing

Fig. 10. Stratigraphic sections of RBT resedimented facies located at Pukekio, “Sheep Terrace” and “Small Spot” (from South toNorth). The base is not exposed as it is covered by the presentsea-level. Facies names are located beside the stratigraphic log for both, resedimented (CH1, CH2, etc), and primary volcanic facies (“A” and “B”). Primary volcanic facies are described inSorrentino et al. (2011).

99L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

topography of Facies CH1, can be produced by deposition from settlingout of suspension, slow-moving sediment clouds or low-concentrationdensity currents (Tucker, 1999; White and Houghton, 2000; Brandand Clarke, 2009 ). However, deposition through suspension andslow-moving sediment clouds would produce a better sorting of thecomponents, particularly in subaqueous environments (Fisher et al.,1998), and the sorting of this facies is poor (Ф = 1.661). Therefore, Fa-cies CH1 is thought to be most likely deposited by low-concentrationdensity flows. The volcanic clasts present in this facies are monomictic(i.e. only one type or category of volcanic clasts) and present

unmodified volcanic textural features (i.e. cuspate shapes, glass shardsthat retain the original shape of the bubble-wall) that represent thetransport of original pyroclasts along a short distance from the volcanicsource (Carey, 2000), in this case, corresponding to early reworkingfrom volcanic Cone II, and deposition b2 km south of it.

In contrast, Facies CH2 is a massive, poorly-sorted mixture of sand-size basaltic volcanic clasts, crystals and fossils fragments set in a siltymatrix, which is partly replaced by cement. The variable thickness,with distinctive hummocks and swales, internally massive structureand poorly-sorted texture, resemble primary volcanic pyroclastic flow

100 L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

deposits (Schmincke et al., 1973; Wilson and Walker, 1982); however,there are no recognised pyroclastic flows within RBT volcanic deposits(Sorrentino et al., 2011), or other contemporaneous volcanic rock inthe Chatham Islands (Campbell et al., 1988). Moreover, pyroclasticflows deposits are unusual in small phreatomagmatic eruptions(Surtseyan-type, Thorarinsson, 1967; Walker, 1973; Kokelaar, 1986;White and Houghton, 2000). Consequently, this calcareous, ash-richsandstone is thought to be a water-supported volcaniclastic mass-flowdeposit generated most likely by a subaqueous explosive eruption ofCone II and emplaced subaqueously. Comparable facies have been illus-trated in the submarine reworked pyroclastic deposit of KrakatauVolca-no in 1883, similar to sample KRK-033A (Mandeville et al., 1996). Thetexturally unmodified monomictic volcanic clasts, in particular, somefluidal clast morphologies (spatter-like), are consistent with earlyerosion of the volcanic edifice, or coeval to depositionwith active volca-nism (Mellors and Sparks, 1991).

The uninterrupted succession of facies CH3 and CH5 is interpretedas being deposited by turbulent subaqueous, granular, mass-flows.This explanation is supported by the progression of structures upwardthrough the section (Fig. 5) produced by a decelerating flow rates(Fisher, 1984). The lower part of the succession is characterisedby normally-graded sandstone followed by laminated reversehydraulically-graded sandstone (higher amount of fossil fragments oflower-density comparedwith the basaltic clasts and crystals) and over-lain by cross-bedded sandstone (Facies CH5). Hence, the whole succes-sion is analogous to well-preserved lower and middle divisions of aBouma Sequence (see McPhie et al., 1993, Fig. 49, pp. 111), lacking thefinely laminated, massive and often bioturbatedmudstone facies typicalof the upper BoumaSequence. It is important to stress that some cautionmust be taken when dealing with transport of bioclastic fragments, be-cause, for example, fossils tend to have a much larger range of sizes andmay create some degree of uncertainty when interpreting environ-ments based purely on grain-size differences. Nevertheless, the men-tioned textural and structural features of these facies can be explainedas a product of low concentration turbidity down-slope currents, andthe lack of upper mud-size Bouma divisions can be interpreted as dueto close proximity with the initiation point or source (see Cas andWright, 1987, pp. 318). Unlike the above mentioned deposits, faciesCH3 and CH5 contain numerous sub-rounded volcanic clasts and lackfluidal or welded clasts; therefore, these facies could indicate post-volcanic erosion, transport and deposition of volcaniclastic debris with-in the limit of storm waves (i.e. ~60–100 m, Fritz and Howells, 1991).

In contrast to facies CH3 and CH5, Facies CH4 is characterised by achannel-like structure, and moderately sorted mudstone, with an ex-tremely high content of texturally immature and compositionally un-modified volcanic clasts (~70%) and extremely low abundance offossils and glauconite. The compositional, structural and textural char-acteristic of this facies resembles some lahar deposits from BridgePoint, Kakanui, New Zealand (Cas and Wright, 1987, Fig. 10.31, pp.328). This implies that the detritus in this facies consists of contempora-neous volcanic debris. Nonetheless, as mentioned in the introduction(i.e. Section 1), distinguishing between primary volcanic deposits andresedimented subaqueous syn-eruptive (or early post-volcanic) de-posits is challenging, particularly when these are not modern or youngdeposits (e.g. Nemeth et al., 2006). Consequently, this facies is consid-ered to have formed by a cohesive flow, where water was an essentialinterstitial medium (Visser, 1983; Postma, 1986); either a primary vol-canic eruption (e.g. syn-eruptive lahar), analogous to some depositsfrom the shallow submarine eruption of Myojinsho, Japan in 1952-1953; (Fiske et al., 1998, Fig. 7) or an epiclastic mud-flow. Facies CH4could, therefore, indicate the re-activation of RBT volcanic activity,specifically re-activation of volcanic Cone II.

The progression of facies towards the top of the RBT resedimentedsuccession is characterised by tractional sedimentary structures. FaciesCH6 is characterised throughout by cross-bedded structures of asym-metrical dunes (indicative of south-westwards currents and consistent

with Cone II as the possible source), probablywith lunate crests. This fa-cies is interpreted as being formed by downstream-migrating bedformsproduced by unidirectional aqueous flows, likely within the fairweatherwave-base (Tucker, 1999, Fig. 2.89, pp. 77). Additionally, the grain mor-phology of clasts or fragments in this facies suggests some degree ofabrasion during transport, particularly the roundness of volcanic clasts(Cas and Wright, 1987; McPhie et al., 1993; White and Houghton,2006). Furthermore, there is no evidence of contemporaneous volcanicactivity (e.g. interbedded ash-layers or fluidal clasts) whilst these clastswere eroded, transported and deposited. Thus, this facies is interpretedas deposited once the volcanic activity had ceased (post-volcanic) inproximal areas to the source (b2 km from volcanic Cone II).

Gradationally overlying Facies CH6 is the calcareous, bedded, ash-rich sandstone, Facies CH7, and whilst cross-bedding characterises theunderlying Facies CH6, Facies CH7 contains abundant planar bedding.Shallowing of the bioclastic and volcaniclastic pile and decreasingslope angle was created by aggradational processes; thus, this facies isalso thought as formed by shallow marine currents, and the depositioncould have occurred as bedload.

Further north, towards the remains of the volcanic edifice, Cone II,Facies CH8 rests unconformably (onlapping contact), on the slopes ofthe cone. There is a continuous upsequence shallowing given by:upsequence increase of in situ fossil assemblages and carbonate content,decreasing glauconitic content and a reduction of mass-flow influx, aswell as abundant sedimentological evidence such as trough cross-bedding. This facies is therefore interpreted as deposited in shallowma-rine volcanic and bioclastic deposits (e.g. Fritz and Howells, 1991). Thecomposition (60%monomictic volcanic clasts) and texture (unmodifiedfragments or clasts and large grain-size) of Facies CH8 support a shortdistance from the source and, therefore, short dispersion of the tephra(Wiesner et al., 1995; Fiske et al., 1998).

An interesting point to mention is the lateral thickness variation ofmost facies (CH1 to CH7), getting thinner towards the remaining volca-nic edifice (Cone II). Moreover, the textural and compositional changes(a decrease in textural maturity, larger grain-size and abundantmonomictic volcanic clasts and glass shards) observed between faciesas getting closer to the volcanic edifice (i.e. CH1-7 vs. CH8, CH26).These changes in the deposition of volcaniclastic debris can be ex-plained as a function of distance to the volcanic source, angle of the vol-canic slope and depth of deposition. These patterns have been explainedin other submarine terrains, such as theMio-PlioceneMisaki Formationin Central Japan (Soh et al., 1989) and theMioceneManukau Subgroup,New Zealand (Allen et al., 2007, and citations therein). In addition,mostfossil assemblages from the lower part of the succession (i.e. facies CH1to CH7) are interpreted as parauthochthonous, whereas most fossilsfrom the upper section of the succession (i.e. facies CH8 and CH26)are inferred to be authochthonous.

The uppermost facies (bedded, calcareous, pyroclast-rich lapillistone,Facies CH26) is located conformably on top of the remains of the volcanicCone II, and is comprised ofmedium-layers (3-10 cm) of abundant, largeand well-preserved fossiliferous detritus (~70%). The size and nature ofthese fossil fragments (i.e. large in situ sponges) could place them inthe boundstone category of the Dunham Scheme. Relatively continuoussedimentation would have been taking place, but likely it would havebeen intermittently abraded by shallowmarine erosion. This facies is in-ferred to have been deposited and accumulated, post-volcanically, by thegrowth of a shallowmarine biota on, or near, the summit of the volcano.The substrate, temperature (Allison, 1973), light saturation, depth (BakerandWeber, 1975), and other marine conditions were favourable to thecolonisation of these previously eroded platforms after themain volcanicactivity had ceased. A review of interesting pioneer investigations aboutthese themes can be found in New (2007). Nonetheless, these faunal re-sponses and their implications for the establishment of shallow marinecommunities in geologic time have not yet been analysed in detail, andthere is much scope for advances in our knowledge of faunal reboundfollowing major volcanic events.

101L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

6. Conclusions

The resedimented facies of the Red Bluff Tuff Formation (upperPalaeocene–lower Eocene) record the subaqueous transport and depo-sition of volcaniclastic and bioclastic debris from the erosion anddegradation of a proximal, entirely submerged Surtseyan volcanicedifice (Cone II). The lowermost distal facies recorded herein were de-posited mainly as water-supported volcanic- or storm-driven massflows (e.g. turbidity currents, mud/debris flows, etc.) of volcaniclasticand bioclastic debris. This is in contrast to the uppermost distal facies,which have characteristics more indicative of tractional sedimentaryprocesses caused by shallow subaqueous currents. Over a geologicallyshort period of time, these stable and wave-eroded volcanic platformswere inhabited by a flourishing marine community of benthic and ses-sile pioneers (corals, bryozoans, molluscs, brachiopods, barnacles,sponges, foraminifera, etc.). This could have happened a number oftimes andmay be interpreted as a reflection of the spectrum of eruptivevolcanic settings and subsequent processes.

Nine resedimented fossiliferous and volcaniclastic facies have beenstudied in detail. This succession exhibits a vertical progression of struc-tures and our interpretations indicate a shallowing upwards sequence.South of the volcanic edifice, the lower section of Pukekio reveals tur-bidity sedimentation in water depths within the storm wave base. Incontrast, the upper section of Pukekio, as well as the section at “SheepTerrace”, is characterised by abundant wave-driven shallow marinetractional structures. Finally, further north, within proximity to thevolcanic edifice, the emplacement of the uppermost facies (CH26) at“Small Spot” is represented by an abundant, diverse, large and wellpreserved in situ fauna of shallow marine sessile organisms.

This research presents the first comprehensive description ofresedimented volcaniclastic and bioclastic facies corresponding to thedegradation of Surtseyan volcanoes on the Chatham Islands. Additional-ly, the outcome of this investigation contributes to a better understand-ing of complex ancient volcano-sedimentary subaqueous terrains. Thismodel of deposition (onlapping/overlapping features on the remainsof a volcanic edifice, a vertical transition of structures from deeper- toshallower-marine environments and preferential colonisation of adiverse biota, including large in situ sessile fauna on the summit) repre-sents an important and novel example to be applied to other ancientsubaqueous volcanic environments.

Acknowledgements

The authors would like to thank Drs. Nils Lenhardt and Brian Jonesfor their review and suggestions to improve this manuscript. We aregrateful with Prof. Ray Cas for his comments during the early prepara-tion of this paper and financial support throughout Leonor Sorrentinopostgraduate candidature. Special thanks go out to the friendly resi-dents of Chatham Island, particularly Valentine and Natasha Croonand their families, and Donna and Terri Tuanui. Sincere thanks tofield-assistants Dr Agustin Cabrera, Ms Deborah Crowley, Dr ChrisConsoli and Dr Steve Poropat for their great enthusiasm and uncondi-tional help on the field. This investigation was funded by an AustralianPostgraduate Award scholarship awarded to Leonor Sorrentino.

References

Allen, S.R., Hayward, B.W., Mathews, E., 2007. A facies model for a submarinevolcaniclastic apron: the Miocene Manukau Subgroup, New Zealand. Geological Soci-ety of America Bulletin 119, 725–742.

Allison, R.C., 1973. Marine paleoclimatology and paleoecology of a Pleistocene inverte-brate fauna from Amchitka Island, Aleutian Islands, Alaska. Palaeogeography,Palaeoclimatology, Palaeoecology 13 (1), 15–48.

Austin, P.M., Sprigg, R.C., Braithwaite, J.C., 1973. Structure and petroleumpotential of east-ern Chatham Rise, New Zealand. American Association of PetroleumGeologists Bulle-tin 57, 477–497.

Baker, P.A.,Weber, J.N., 1975. Coral growth rate: variation with depth. Physics of the Earthand Planetary Interiors 10 (2), 135–139.

Bennett, F.D., 1972. Shallow Submarine Volcanism. Journal of Geophysical Research 77(29), 5755–5759.

Blott, S.J., Pye, K., 2001. GRADISTAT: a grain size distribution and statistics package for theanalysis of unconsolidated sediments. Earth Surface Processes and Landforms 26,1237–1248.

Brand, B.D., Clarke, A.B., 2009. The architecture, eruptive history, and evolution of theTable Rock Complex, Oregon: from a Surtseyan to an energeticmaar eruption. Journalof Volcanology and Geothermal Research 180 (2), 203–224.

Brown, S.J.A., Smith, R.T., Cole, J.W., Houghton, B., 1994. Compositional and textural char-acteristics of the strombolian and surtseyan K-Trig basalts, Taupo Volcanic Centre,New Zealand: implications for eruption dynamics. New Zealand Journal of Geologyand Geophysics 37, 113–126.

Busby-Spera, C.J., 1988. Evolution of a Middle Jurassic back-arc basin, Cedros Island, BajaCalifornia: evidence from amarine volcaniclastic apron. Geological Society of AmericaBulletin 100 (2), 218–233.

Campbell, H., Hutching, G., 2007. In search of ancient New Zealand. Penguin Books andGNS Science, Lower Hutt, New Zealand (239 pp.).

Campbell, H.J., et al., 1988. Cretaceous–Cenozoic lithostratigraphy of the Chatham Islands.Journal of the Royal Society of New Zealand 18 (3), 285–308.

Campbell, H.J., et al., 1993. Cretaceous–Cenozoic geology and biostratigraphy of the ChathamIslands, New Zealand. Institute of Geological and Nuclear Sciences Monograph 2, 269.

Cantelli, A., Johnson, S., White, J.D.L., Parker, G., 2008. Sediment Sorting in the Deposits ofTurbidity Currents Created by ExperimentalModeling of Explosive Subaqueous Erup-tions. The Journal of Geology 116 (1), 76–93.

Carey, S., 2000. Volcaniclastic sedimentation around island arcs, Encyclopedia of Volca-noes. Academic Press (1442 pp.).

Carey, S., Sigurdsson, H., Mandeville, C., Bronto, S., 1996. Pyroclastic flows and surges overwater: an example from the 1883 Krakatau eruption. Bulletin of Volcanology 57 (7),493–511.

Cas, R.A.F., Wright, J.V., 1987. Volcanic Successions. Modern and Ancient. Chapman andHall, London (528 pp.).

Cas, R.A.F., Porritt, L., Pittari, A., Hayman, P.C., 2009. A practical guide to terminology forkimberlite facies: a systematic progression from descriptive to genetic, including apocket guide. Lithos 112 (Supplement 1), 183–190.

Consoli, C.P., Stilwell, J.D., 2011. Palaeontology of the Chatham Islands, SW Pacific−a re-view. Alcheringa An Australasian Journal of Palaeontology 35 (2), 285–301.

Eagles, G., Gohl, K., Larter, R.D., 2004. High-resolution animated tectonic reconstruction ofthe South Pacific and West Antarctic Margin. Geochemistry, Geophysics, Geosystems5 (7).

Fisher, R.V., 1984. Submarine volcaniclastic rocks. Geological Society, London, Special Pub-lications 16 (1), 5–27.

Fisher, R.V., Schmincke, H.-U., 1984. Pyroclastic Rocks. Springer–Vergal, New York (472 pp.).Fisher, R.V., Cashman, K.V., Shibata, A., Watanabe, K., 1998. Tephra dispersal from

Myojinsho, Japan, during its shallow submarine eruption of 1952–1953. Bulletin ofVolcanology 59, 262–275.

Fiske, R.S., Cashman, K.V., Shibata, A., Watanbe, K., 1998. Tephra dispersal fromMyojinsho, Japan, during its shallow submarine eruption of 1952–1953. Bulletin ofVolcanology 59, 262–275.

Fritz, W.J., Howells, M.F., 1991. A shallow marine volcaniclastic facies model: an examplefrom sedimentary rocks bounding the subaqueously welded Ordovician Garth Tuff,North Wales, U.K. Sedimentary Geology 74 (1–4), 217–240.

Grindley, G.W., Adams, C.J., Lumb, J.T., Watters, W.A., 1977. Paleomagnetism, K–Ar datingand tectonic interpretation of Upper Cretaceous and Cenozoic volcanic rocks of Chat-ham Islands, New Zealand. New Zealand Journal of Geology and Geophysics 20 (3),425–467.

Hay, R.F., Mutch, A.R., Watters, W.A., 1970. The geology of the Chatham Islands, NewZealand. Geological Survey Bulletin 83, 1–86.

Herzer, R.H., Wood, R.A., 1988. Tectonic history, sedimentation, and changes in relative sealevel, Chatham Rise, New Zealand. New Zealand Geological Survey Record 29, 1–22.

Houghton, B.F., Landis, C.A., 1989. Sedimentation and volcanism in a Permian arc-relatedbasin, southern New Zealand. Bulletin of Volcanology 51 (6), 433–450.

Kokelaar, B.P., 1986. Magma-water interactions in subaqueous and emergent basaltic vol-canism. Bulletin of Volcanology 48, 275–289.

Kokelaar, B.P., Durant, G.P., 1983. The submarine eruption and erosion of Surtla (Surtsey),Iceland. Journal of Volcanology and Geothermal Research 19 (3–4), 239–246.

Laird, M.G., Bradshaw, J.D., 2004. The Break-up of a Long-term Relationship: the CretaceousSeparation of New Zealand from Gondwana. Gondwana Research 7 (1), 273–286.

Larter, R.D., Cunningham, A.P., Barker, P.F., Gohl, K., Nitsche, F.O., 2002. Tectonic evolutionof the Pacific margin of Antarctica 1. Late Cretaceous tectonic reconstructions. Journalof Geophysical Research 107 (B12), 2345.

Mandeville, C., Carey, S., Sigurdsson, H., 1996. Sedimentology of the Krakatau 1883 sub-marine pyroclastic deposits. Bulletin of Volcanology 57 (7), 512–529.

Mays, C., Stilwell, J.D., 2012. Judging an acritarch by its cover: the taxonomic implicationsof Introvertocystis rangiaotea gen. et sp. nov. from the Late Cretaceous (Cenomanian–Turonian) of the Chatham Islands, New Zealand. Palynology 36, 180–190.

Mays, C., Stilwell, J.D., 2013. Pollen and spore biostratigraphy of the mid-CretaceousTupuangi Formation, Chatham Islands, New Zealand. Review of Palaeobotany andPalynology 192, 79–102.

McPhie, J., Doyle, M., Allen, R., 1993. Volcanic Textures: A Guide to the Interpretation ofTextures in Volcanic Rocks. Centre for Ore Deposit and Exploration Studies, Universi-ty of Tasmania, Hobart (196 pp.).

Mellors, R.A., Sparks, R.S.J., 1991. Spatter-rich pyroclastic flow deposits on Santorini,Greece. Bulletin of Volcanology 53 (5), 327–342.

Molnar, P., 1975. Magnetic Anomalies, Bathymetry and the Tectonic Evolution of theSouth Pacific since the Late Cretaceous. The Geophysical Journal of the Royal Astro-nomical Society 40 (3), 383–420.

102 L. Sorrentino et al. / Sedimentary Geology 300 (2014) 86–102

Nemeth, K., Cronin, S.J., Charley, D.T., Harrison, M.J., Garae, E., 2006. Exploding lakes inVanuatu-“ Surtseyan-style” eruptions witnessed on Ambae Island. Episodes 29 (2),87–92.

New, T.R. (Ed.), 2007. Island colonization: the origin and development of island commu-nities. Ecological Reviews. Cambridge University Press (287 pp.).

Odin, G.S., Matter, A., 1981. De glauconiarium origine. Sedimentology 28, 611–641.Pole, M., Philippe, M., 2010. Cretaceous plant fossils of Pitt Island, the Chatham group,

New Zealand. Alcheringa 34, 231–263.Postma, G., 1986. Classification for sediment gravity-flow deposits based on flow condi-

tions during sedimentation. Geology 14 (4), 291–294.Schmincke, H.U., Fisher, R.V.,Waters, A.C., 1973. Antidune and chute and pool structures in

the base surge deposits of the Laacher See area, Germany. Sedimentology 20 (4),553–574.

Schneider, C.A., Rasband, W.S., Eliceiri, K.W., 2012. NIH Image to ImageJ: 25 years ofimage analysis. Nature Methods 9 (7), 671–675.

Soh, W., et al., 1989. Submarine depositational processes for volcaniclastic sediments inthe Mio-Pliocene Misaki formation, Miura Group, Central Japan. Sedimentary Faciesin the Active Plate Margin.Terra Scientific Publishing Company (TERRAPUB), Tokyo619–630.

Sorrentino, L., Cas, R.A.F., Stilwell, J.D., 2011. Evolution and facies architecture of PaleogeneSurtseyan volcanoes on Chatham Islands, New Zealand, Southwest Pacific Ocean.Journal of Volcanology and Geothermal Research 202 (1–2), 1–21.

Stilwell, J.D., 1997. Tectonic and palaeobiographic significance of the Chatham Islands,South Pacific, Late Cretaceous fauna. Palaeogeography, Palaeoclimatology, Palaeo-ecology 136, 97–119.

Stilwell, J.D., Consoli, C.P., 2012. Tectono-stratigraphic history of the Chatham Islands, SWPacific-The emergence, flooding and reappearance of eastern ‘Zealandia’. Proceedingsof the Geologists Association 123 (1), 170–181.

Stilwell, J.D., et al., 2006. Dinosaur sanctuary on the Chatham Islands, Southwest Pacific:first record of theropods from the K–T boundary Takatika Grit. Palaeogeography,Palaeoclimatology, Palaeoecology 230, 243–250.

Sutherland, R., 1995. The Australia–Pacific boundary and Cenozoic plate motions in theSW Pacific: some constraints from Geosat data. Tectonics 14 (4), 819–831.

Sutherland, R., 1999. Basement geology and tectonic development of the greater NewZealand region: an interpretation from regional magnetic data. Tectonophysics 308(3), 341–362.

Thorarinsson, S., 1967. Surtsey. The New Island in the North Atlantic. The Viking Press,New York 1–47.

Thorarinsson, S., Einarsson, T., Sigvaldason, G., Ellison, G., 1964. The submarine eruptionoff the Vestmann Islands. Bulletin of Volcanology 27, 435–445.

Timm, C., et al., 2010. Temporal and geochemical evolution of the Cenozoic intraplate vol-canism of Zealandia. Earth Science Reviews 98 (1–2), 38–64.

Trofimovs, J., et al., 2006. Submarine pyroclastic deposits formed at the Soufriere Hills vol-cano, Montserrat (1995–2003): what happens when pyroclastic flows enter theocean? Geology 34 (7), 549–552.

Trofimovs, J., Sparks, R.S.J., Talling, P.J., 2008. Anatomy of a submarine pyroclastic flow andassociated turbidity current: July 2003 dome collapse, Soufrière Hills volcano, Mont-serrat, West Indies. Sedimentology 55 (3), 617–634.

Tucker, M.E., 1999. Sedimentary Petrology. An Introduction to the Origin of SedimentaryRocks Geoscience Texts. Blackwell Science London (253 pp.).

Tulloch, A.J., et al., 2009. Cretaceous felsic volcanism in New Zealand and Lord Howe Rise(Zealandia) as a precursor to final Gondwana break-up. Geological Society, London,Special Publications 321 (1), 89–118.

Visser, J.N.J., 1983. Submarine debris flow deposits from the Upper CarboniferousDwyka Tillite Formation in the Kalahari Basin, South Africa. Sedimentology 30(4), 511–523.

Walker, G.P.L., 1973. Explosive volcanic eruptions—a new classification scheme. Interna-tional Journal of Earth Sciences 62 (2), 431–446.

White, J.D.L., Busby-Spera, C.J., 1987. Deepmarine arc apron deposits and syndepositionalmagmatism in the Alisitos group at Punta Cono, Baja California, Mexico. Sedimentol-ogy 34 (5), 911–927.

White, J.D.L., Houghton, B., 2000. Surtseyan and related phreatomagmatic eruptions,Encyclopedia of Volcanoes. Academic Press, San Diego, U.S.A.(1442 pp.).

White, J.D.L., Houghton, B., 2006. Primary volcaniclastic rocks. Geology 34, 677–680.Wiesner, M.G., Wang, Y., Zheng, L., 1995. Fallout of volcanic ash to the deep South China

Sea induced by the 1991 eruption of Mount Pinatubo (Philippines). Geology 23 (10),885–888.

Wilson, C.J.N., Walker, G.P.L., 1982. Ignimbrite depositional facies: the anatomy of a pyro-clastic flow. Journal of the Geological Society 139 (5), 581–592.

Wood, R.A., Andrews, P.B., Herzer, R.H., et al., 1989. Cretaceous and Cenozoic geology ofthe Chatham Rise Region, South Island, New Zealand. New Zealand Geological SurveyBasin Studies 3, 76.