Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Kimberlite volcanic facies and eruption in the Buffalo Head Hills, Alberta (Canada) Boyer, Liane Patricia 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2005-0014.pdf [ 25.33MB ]
JSON: 831-1.0052751.json
JSON-LD: 831-1.0052751-ld.json
RDF/XML (Pretty): 831-1.0052751-rdf.xml
RDF/JSON: 831-1.0052751-rdf.json
Turtle: 831-1.0052751-turtle.txt
N-Triples: 831-1.0052751-rdf-ntriples.txt
Original Record: 831-1.0052751-source.json
Full Text

Full Text

K L M B E R L I T E V O L C A N I C FACFFIS A N D ERUPTION IN T H E B U F F A L O H E A D H L L L S , A L B E R T A ( C A N A D A ) by L I A N E PATRICIA B O Y E R B.Sc.Eng., (Honours), Queen's University, 2002 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L M E N T OF T H E R E Q U I R E M E N T S FOR T H E D E G R E E OF M A S T E R OF SCIENCE in T H E F A C U L T Y OF G R A D U A T E STUDIES G E O L O G I C A L SCIENCES T H E UNIVERSITY OF BRITISH C O L U M B I A January 2005 © Liane Patricia Boyer, 2005 Abstract An analysis of volcanic facies in kimberlites from the Buffalo Head Hills (BHH) of Northern Alberta, Canada, is undertaken to discover the mechanisms of kimberlite eruption and deposition in this region. Thirty-eight kimberlites have been discovered in the BHH to date; six were chosen for this study. Kimberlite bodies K6, K l 1, K252, K281, K296 and K300 were selected to represent the range of deposits found in BHH with the object of constraining the eruptive style of these kimberlites. The lithofacies observed in bodies K6 and K l 1 have characteristics consistent with deposition through pyroclastic fall and synvolcanic slump. Rare accretionary pyroclasts, abundant juvenile pyroclasts, bomb sags, and weak bedding, attest to a pyroclastic origin and the geometry of the deposits suggest proximal deposition within a crater. Body K281 lithofacies are characterized by abundant accretionary pyroclasts and alternating intervals of pyroclastic surge and pyroclastic fall deposits. Similar to bodies K6 and K l 1, the stratigraphy of body K281 suggests deposition within a crater, however the better developed sorting and structures suggest a less proximal, crater edge location. The deposits and geometry of body K296 include massive pyroclastic fall occurring with a crater, and finely bedded, relatively well-sorted pyroclastic surge deposits occurring in an outer tephra ring. The deposits of body K300 are also surge dominated with a geometry that implies a tephra ring location, however no intra-crater deposits have been discovered in this body. Finally, outside of the tephra ring, distal pyroclastic fall deposits and reworked pyroclastic kimberlite occur, and these are recorded in the deposits of body K252. The characteristics of each of the bodies indicate similar eruption and depositional mechanisms. Information from each of these bodies can be combined to create a composite model of an idealized kimberlite in this region. According to this model the kimberlite is ii composed of a pyroclastic surge-dominated tephra ring surrounding a crater that is partially infilled with pyroclastic material, and subsequently with material reworked from the tephra ring. These deposits and clast types, particularly the presence of pyroclastic surge deposits and accretionary pyroclasts, are good indications that phreatomagmatism was an important mechanism of kimberlite eruption in this region. However, the high volatile contents traditionally associated with kimberlite, in concert with an alteration assemblage indicating abundant C O z , suggest that exsolution driven explosive volcanism is likely the dominant mechanism in kimberlite emplacement. Frequently these two end-member models have been considered mutually exclusive, however deposits in B H H require both mechanisms. The model proposed includes the concept of magmatic fragmentation through exsolution, nucleation and expansion of volatiles, as well as the incorporation of external water or wet sediments resulting in phreatomagmatic interactions. In this model phreatomagmatism is the result of turbulent mixing of a dispersed flow of magma and gases with surrounding wet sediments. This model accounts for the range of features observed in Buffalo Head Hills kimberlites, while allowing for the presence of reasonable volatile contents within the kimberlite magma. iii Table of Contents ABSTRACT ii T A B L E OF CONTENTS iv LIST OF FIGURES vi LIST OF TABLES ix LIST OF PLATES ix ACKNOWLEDGEMENTS x CHAPTER 1 OVERVIEW 1 CHAPTER 2 REGIONAL G E O L O G Y AND PREVIOUS WORK 3 2.1 INTRODUCTION 3 2.2 GEOLOGIC FRAMEWORK 7 2.2.1 PRECAMBRIAN 7 2.2.2 DEVONIAN 9 2.2.3 CRETACEOUS AND TERTIARY 9 2.2.4 QUATERNARY : 11 2.3 CLASSIFICATION OF ROCKS AS KIMBERLITE 12 CHAPTER 3 VOLCANIC FACffiS 14 3.1 INTRODUCTION 14 3.1.1 DEPOSIT CHARACTERISTICS 15 3.2 VOLCANOLOGICAL CLASSIFICATION OF KLMBERLITES : 16 3.3 OVERVIEW, OF KIMBERLITES .; 19 3.3.1 PALEO-ENVIRONMENT 19 3.3.2 KiMBERLrre GEOMETRY 20 3.3.3 CLAST POPULATION NOMENCLATURE 20 3.3.4 DATA PRESENTATION 25 3.4 DISCUSSION OF BODIES '. 26 iv 3.4.1 B O D Y K 6 .• 26 3.4.2 B O D Y K l 1 39 3.4.3 B O D Y K 2 5 2 51 3.4.4 B O D Y K 2 8 1 '. 74 3.4.5 B O D Y K 2 9 6 87 3.4.6 B O D Y K 3 0 0 103 3.5 S U M M A R Y 116 C H A P T E R 4 C O M P O S I T E V O L C A N O L O G I C A L M O D E L 119 4.1 RATIONALE FOR THE COMPOSITE MODEL 119 4.2 COMPOSITE VOLCANOLOGICAL MODEL 119 4.3 R E L E V A N C E OF THE VOLCANOGENIC MODEL TO ERUPTIVE STYLE 123 C H A P T E R 5 E R U P T I V E M E C H A N I S M S 127 5.1 T H E EXSOLUTION MODEL 133 5.2 T H E PHREATOMAGMATIC MODEL 135 5.3 LIMITATIONS OF CURRENT MODELS 137 5.4 ERUPTIVE MODEL FOR B U F F A L O H E A D HILLS KIMBERLITE 139 C H A P T E R 6 C O N C L U S I O N ; 145 R E F E R E N C E S 147 v List of Figures FIGURE 2.1: BEDROCK GEOLOGY OF THE BUFFALO HEAD HILLS 4 FIGURE 2.2: DISTRIBUTION OF KIMBERLITES WITHIN THE BUFFALO H E A D HILLS 5 FIGURE 2.3: SIMPLIFIED PHANEROZOIC STRATIGRAPHY OF THE B U F F A L O H E A D HILLS 6 FIGURE 2.4: PRECAMBRIAN BASEMENT DOMAINS OF A L B E R T A 8 FIGURE 3.1: VARIATIONS IN CROSS-SECTIONAL GEOMETRY OF KIMBERLITE PIPES 17 FIGURE 3.2: KIMBERLITE CLASSIFICATION SCHEME OF FIELD A N D SCOTT SMITH 18 FIGURE 3.3:' SURFACE EXPRESSIONS OF BODIES K6 A N D K252 •. 27 FIGURE 3.4: KIMBERLITE INTERSECTIONS IN DRILLHOLES THROUGH BODY K6 28 FIGURE 3.5: STRATIGRAPHIC SECTION FOR DRILLHOLE K6-13 30 FIGURE 3.6: LITHOFACIES OF BODY K6 32 FIGURE 3.7: PROPOSED GEOMETRY A N D DEPOSITS OF BODY K6 35 FIGURE 3.8: SURFACE EXPRESSION A N D KIMBERLITE INTERSECTIONS OF BODY Kl 1 40 FIGURE 3.9: STRATIGRAPHIC SECTION FOR DRILLHOLE Kl 1-01 41 FIGURE 3.10: JUVENILE PYROCLASTS IN BODY Kl 1 44 FIGURE 3.11: OLIVINE ALTERATION IN BODY Kl 1 45 FIGURE 3.12: OLIVINE PHENOCRYSTS IN BODY Kl 1 46 FIGURE 3.13: PROPOSED GEOMETRY A N D DEPOSITS OF BODY Kl 1 49 FIGURE 3.14: KIMBERLITE INTERSECTIONS IN DRILLHOLES THROUGH BODY K252 52 FIGURE 3.15: STRATIGRAPHIC SECTION FOR DRILLHOLE K252-01 53 FIGURE 3.16: STRATIGRAPHIC SECTION FOR DRILLHOLE K252-03 54 FIGURE 3.17: STRATIGRAPHIC SECTION FOR DRILLHOLE K252-08 55 FIGURE 3.18: STRATIGRAPHIC SECTION FOR DRILLHOLE K252-09 56 FIGURE 3.19: STRATIGRAPHIC SECTION FOR DRILLHOLE K252-12 57 FIGURE 3.20: PROPOSED GEOMETRY A N D DEPOSITS OF BODY K252 59 FIGURE 3.21: LITHOFACIES 1 A N D 2 OF BODY K252 60 FIGURE 3.22: LITHOFACIES 3 A N D 4 OF BODY K252 61 FIGURE 3.23: JUVENILE PYROCLASTS OF BODY K252 64 vi FIGURE 3.24: ACCRETIONARY PYROCLASTS OF BODY K252 66 FIGURE 3.25: SURFACE EXPRESSION A N D KIMBERLITE INTERSECTIONS OF BODY K281 75 FIGURE 3.26: STRATIGRAPHIC SECTION FOR DRILLHOLE K281-01 76 FIGURE 3.27: LITHOFACIES OF BODY K281 77 FIGURE 3.28: JUVENILE PYROCLASTS OF BODY K281 80 FIGURE 3.29: ACCRETIONARY PYROCLASTS OF BODY K281 81 FIGURE 3.30: PROPOSED GEOMETRY AND DEPOSITS OF BODY K281 84 FIGURE 3.31: SURFACE EXPRESSION A N D KIMBERLTTE INTERSECTIONS OF BODY K296 88 FIGURE 3.32: STRATIGRAPHIC SECTION FOR DRILLHOLE K296-01 89 FIGURE 3.33: STRATIGRAPHIC SECTION FOR DRILLHOLE K296-02 90 FIGURE 3.34: STRATIGRAPHIC SECTION FOR DRILLHOLE K296-03 91 FIGURE 3.35: PROPOSED GEOMETRY AND DEPOSITS OF BODY K296 93 FIGURE 3.36: LITHOFACIES 1 OFBODYK296 94 FIGURE 3.37: LITHOFACIES 2 OF BODY K296 95 FIGURE 3.38: LITHOFACIES 3 OF BODY K296 96 FIGURE 3.39: LITHOFACIES 4 OF BODY K296 98 FIGURE 3.40: SURFACE EXPRESSION AND KIMBERLITE INTERSECTIONS OF BODY K300 104 FIGURE 3.41: STRATIGRAPHIC SECTION FOR DRILLHOLE K300-01 106 FIGURE 3.42: STRATIGRAPHIC SECTION FOR DRILLHOLE K300-02 107 FIGURE 3.43: STRATIGRAPHIC SECTION FOR DRILLHOLE K300-03 108 FIGURE 3.44: PROPOSED GEOMETRY AND DEPOsrrs OF BODY K300 109 FIGURE 3.45: LITHOFACIES OF BODY K300 110 FIGURE 3.46: CLAST POPULATION OF BODY K300 112 FIGURE 4.1: COMPOSITE VOLCANOLOGICAL MODEL OF A BUFFALO HEAD HILLS KIMBERLITE 120 FIGURE 4.2: 3-D MODEL OF A BUFFALO HEAD HILLS KIMBERLITE 122 FIGURE 5.1: KIMBERLITE MORPHOLOGY: EXSOLUTION MODEL 128 FIGURE 5.2: E M P L A C E M E N T OF KIMBERLITE: EXSOLUTION MODEL 129 FIGURE 5.3: KIMBERLITE MORPHOLOGY: PHREATOMAGMATIC MODEL 130 FIGURE 5.4: E M P L A C E M E N T OF KIMBERLITE: PHREATOMAGMATIC MODEL 131 vii FIGURE 5.5: E M P L A C E M E N T MECHANISMS OF KIMBERLITE: PHREATOMAGMATIC MODEL FIGURE 5.6: PROPOSED MODEL FOR KIMBERLITE ERUPTION IN THE BUFFALO HEAD HILLS viii List of Tables T A B L E 2.1: P A L Y N O L O G Y OF MUDSTONES ADJACENT TO KIMBERLITE IN THE BUFFALO H E A D HILLS 10 T A B L E 3.1: DISTINGUISHING FEATURES OF JUVENILE, ACCRETIONARY A N D AUTOLITHIC PYROCLASTS 22 T A B L E 3.4: CHARACTERISTICS OF KIMBERLITE DEPOSITS IN BODY K6 29 T A B L E 3.5: CHARACTERISTICS OF KIMBERLITE DEPOSITS IN BODY Kl 1 42 T A B L E 3.6: CHARACTERISTICS OF KIMBERLITE DEPOSITS IN BODY K252 68 T A B L E 3.7: CHARACTERISTICS OF KIMBERLITE DEPOSITS IN BODY K281 78 T A B L E 3.8: CHARACTERISTICS OF KIMBERLITE DEPOSITS \N BODY K296 '. 92 T A B L E 3.9: CHARACTERISTICS OF KIMBERLITE DEPOSITS IN BODY K300 105 T A B L E 3.10: S U M M A R Y OF CHARACTERISTICS OF BODIES K6, K11, K252, K281, K296 A N D K300 117 List of Plates P L A T E 1: DEPOSIT TYPES TYPICAL OF VOLCANIC ENVIRONMENTS A N D THEIR CHARACTERISTICS ix Acknowledgements I would like to thank my supervisor, Dick Tosdal, and co-supervisors Tom McCandless and Kelly Russell. Your contributions to my thesis were all very different, but all very appreciated. I feel very fortunate to have had such great supervisors. Financial support for this project was through a grant set up by the Mineral Deposit Research Unit that included contributions from the NSERC and Ashton Mining of Canada Inc. with their joint venture partners Pure Gold Resources and Encana. Special thanks to all employees of Ashton Mining of Canada for their assistance in many different aspects of my project. I am grateful to the DeBeers Petrography unit for introducing me to the world of kimberlite petrology and to J.P Zonneveld at the GSC for answering my sedimentological questions. Thanks to Maya Kopylova whose revisions improved the quality of this document. Also thanks to Arne Toma and Ken Hickey for their expertise and patience in computer/printing issues, and to James Scoates for the use of an excellent microscope. A l l of my fellow grad students, as well as many undergrads, and professors, have made my M.Sc. a great experience, and I am grateful to them all. In particular, my wonderful office mates, Maggie Harder, Nathalie Lefebvre, Dianne Mitchinson, and honorary office mate Heidi Annell, deserve huge thanks for providing support in so many ways. , Lastly I would like to thank my family, and in particular Dan, Bear, Elliott and Teriyaki, who have been through it all, and are still around. Thanks. x Chapter 1: Overview Eruptive mechanisms for kimberlite are currently the subject of considerable debate. On one side are models invoking the exsolution of C 0 2 from magma at shallow depths leading to the explosive eruption of kimberlite onto the Earth's surface (Clement, 1982). The basis for this model lies largely in observations of the deeper parts of kimberlite that are now near the surface due to erosion and are exploited largely in South Africa. On the other side are models invoking phreatomagmatic processes (Lorenz, 1986). Such models derive from correlations between kimberlites and maar volcanoes and the association of maars with phreatomagmatic processes. In this thesis, new data and interpretations are presented that support both an exsolution and phreatomagmatic driven component to kimberlite eruption, but by different mechanisms from*the traditional models. This study is composed of a volcanologic study of the Buffalo Head Hills (BHH) kimberlite occurrences that incorporates aspects such as country rock geology, kimberlite geometry, deposit types and occurrence, as well as clast types and distribution. From this study a model for an 'ideal' B H H kimberlite is created, and the eruptive mechanisms necessary to generate the physical aspects are discussed. The Buffalo Head Hills kimberlite field is located in north central Alberta, and is one of several significant new kimberlite provinces discovered in Canada within the last twenty years. Thirty-eight kimberlites have been discovered to date, and more are likely to be found beneath the Quaternary glacial deposits. The volcanology of these bodies is largely unknown as they rarely outcrop and none have been excavated. The material available is drill core from diamond exploration drilling, and it is these rocks that form the data set for this study. Preliminary work prior to this study (Carlson et al. 1999) suggests that these bodies are largely volcaniclastic and thus represent the near surface and/or extrusive parts of a kimberlite volcano. As a result 1 volcanic facies analysis of different parts of a selected subset of the 38 currently known kimberlites in the B H H is completed herein to form the basis for the volcanologic model presented. The volcanological analysis of the kimberlites is undertaken in three parts. In chapter 2, the regional geologic setting of the kimberlites is reviewed, as is the state of knowledge about these bodies prior to this study. A n overview of the geologic sequence through which kimberlite ascended is included, as are ages of selected kimberlites in this region. The distribution and known locations of kimberlite is discussed, as well as the classification of these rocks as kimberlite. Chapter 3 comprises a volcanic facies investigation of six bodies in the Buffalo Head Hills region. This investigation involves the characterization of each body, from diamond drill core, and the creation of a model for the physical aspects and eruptive processes inferred for each body. In Chapter 4 a schematic model for an ideal B H H kimberlite is presented, that amalgamates the features discussed in Chapter 3, as well as incorporating information about this volcanic province presented in Chapter 2. The model is then interrogated for constraints on the eruption mechanisms. Traditionally two models have been proposed for kimberlite volcanism. The first asserts that kimberlite ascent and eruption is driven by the exsolution of volatiles, dominantly C O z . The second model proposes that kimberlites erupt phreatomagmatically, and that the presence of large amounts of C 0 2 is not necessary for the ascent and eruption of kimberlites. The two models are herein critically reviewed and discussed in reference to the features observed, and model constructed, for B H H kimberlites. 2 Chapter 2: Regional Geology and Previous Work 2.1 Introduction The kimberlites that make up the Buffalo Head Hills kimberlite province occur in north central Alberta in the Buffalo Head Hills region (FIGURE 2.1, 2.2). The Buffalo Head Hills region is an area of topographic relief within the Interior Plains of north-central Alberta where maximum elevations on the crests of the hills are on the order of 770 m. (Wall and Singh, 1975). Kimberlites were discovered in the Buffalo Head Hills (BHH), in January of 1997 by Ashton Mining of Canada Inc. (Carlson et al., 1999; Ashton Mining of Canada Inc., 1997). The discovery resulted from oil and gas exploration by the Alberta Energy Company (now Encana) to investigate the structure of the Precambrian basement and Phanerozoic stratigraphy in this region (Skelton et al. 2003). Anomalous steep reflectors identified in seismic data were found to correlate with anomalies noted from aeromagnetic data. Drilling of these anomalies resulted in the identification of kimberlite. To date, 38 kimberlites have been discovered in the B H H region; 26 are proven diamondiferous (Ashton mining of Canada Inc., 2003a,b) and host significant (up to 55 carats per hundred tons) quantities of diamond (Ashton Mining of Canada Inc., 2001). Extensive Pleistocene glacial deposits, which range in thickness from a few metres to several hundred metres (Pawlowicz and Fenton 1995), obscure the location, distribution and geometry of individual kimberlites. The area is also variably covered by muskeg and dense forest. Despite this lack of outcrop, a considerable amount of information on the subsurface geology is available as a result of extensive oil and gas exploration, including drilling. Beneath the glacial deposits are ~1600 m of Phanerozoic strata overlying Precambrian basement (FIGURE 2.3; Mossop and Shetson, 1994; Carlson et al. 1999). 3 Explanation Ksh =Shaftesbury Formation Ks = Smoky Group Kd = Dunvegan Formation Kib = Kimberlite intrusive body ^ = Location of kimberlite Figure 2.1: Bedrock geology of Alberta projected beneath overlying Quaternary rocks and sediments. Modified from Hamilton et al. (1998). Figure 2.2: Distribution of kimberlites within the Buffalo Head Hills Claim Block. Kimberlites K6, K11, K252, K281, K296 and K300 are studied in this thesis. 5 Quaternary 0-150 m Cretaceous -325 m Devonian -1300 m Precambrian Basement Time Paleocene 66.4 Maastrichtian 74.5 => o Campanian LU 84.0 87.5 Santonian Coniadan U Turanian 91 97.5 Cenomanian Albian 113 Aptian Stratigraphy in BHH region Removed by glaciation Kaskapau Dunvegan Sh I E BF,FS,Wg EES Spirit River Bluesky /Gething Kimberlite intervals 64.2 -U/Pb (dyke) 83.2 87.6 U/Pb 89 Palynology 97 - 98.9 Paleo-environment from palynology Open Marine Marine Anoxic Estuarine Figure2.3: Simplified Phanerozoic stratigraphy of the Buffalo Head Hills. Cretaceous section with simplified stratigraphy and intervals of kimberlite volcanism is enlarged. Compiled from Dufresne et al. (2001), Alberta core data sheet, Carlson et al. (1999), Ashton Mining of Canada Internal Report (2000). Abbreviations: Sh= Shaftesbury, BF = Belle Fourche, FS = Fish scales, Wg= Westgate, PR = Peace River, P/C = Paddy/Cadotte, H= Harmon. 6 2.2 Geologic Framework 2.2.1 Precambrian T h e crystalline basement in the B H H area o f Alberta does not outcrop, and is presently subject to confl icting interpretations. There are two proposed divis ion schemes for the basement in the Buffa lo H e a d H i l l s area. A c c o r d i n g to Ross (1990) Ross et al. (1991, and Ross and Eaton (1997), the Buf fa lo H e a d H i l l s lie within the geophysically defined Buffa lo H e a d Terrane (FIGURE 2.4A). T h i s Terrane is a broad region o f generally positive, north-trending, arcuate aerbmagnetic anomalies that constitute a convex-westward arc-shaped domain that is inferred to be bounded by structural discontinuities. Ross (1997) further proposed that the Buffa lo H e a d Terrane is composed o f Proterozoic rocks formed between 2.0 and 2.32 G a that have been affected by a widespread younger (1.9-2.0 G a ) thermal magmatic event. However , Ross (1990) also states that S m - N d data (Theriault and Ross 1991) indicate that the lower crustal foundation o f each o f these Proterozoic arcs is Archean material. Ross (1990, 2002) interprets this and other domains in northern Alberta as representing A r c h e a n continental blocks separated by oceanic lithosphere which , through progressive convergence and subduction o f oceanic crust, led to the formation o f over ly ing lower Proterozoic magmatic rocks as magmatic arcs. In contrast, Burwash et al. (2000) util ized aeromagnetic data, metamorphic facies and seismic discontinuities to propose different tectonometamorphic divisions for the Precambrian basement o f Alber ta (FIGURE 2.4B). In this model , the Buffa lo H e a d H i l l s lie within the R e d Earth Granulite D o m a i n o f the Athabasca Polymetamorphic Terrane. B u r w a s h et al. (2000) further suggest that this terrane is Archean-age crustal material that was reworked during the Paleoproterozoic and that the Athabasca Polymetamorphic Terrane represents the southwestward extension o f the pre-Taltson basement. Burwash et al. (2000) state that this terrane has a complex 7 Figure 2.4: Models of the Precambrian basement terrane of Alberta. A. Tectonic d ivisions of Ross et al.(1991) B. Tectonic divisions of Burwash etal. (2000). 8 lower Proterozoic history as they obtained U-Pb ages between 1900 to 2400 M a as well as Archean Nd-model ages (Villeneuve et al., 1993). Kimberlites in the B H H lie within the Red Earth Granulite Domain, which is an elliptical area of granulite-facies metamorphism around Red Earth Creek. The significance of the Red Earth Granulite Domain is unknown. 2.2.2 Devonian Unconformably overlying the Precambrian basement is a middle to upper Devonian sequence, comprising carbonate, clastic, and evaporitic rock, that attains a thickness up to 1300 m (FIGURE 2.3). This sequence includes the Granite Wash, Keg River, and Slave Point Formations and the Wabumum Group, which have been exploited for oil and gas in this region (Carlson et al., 1999; Mossop and Shetson, 1994). 2.2.3 Cretaceous and Tertiary Cretaceous sedimentary rocks unconformably overlie the Devonian sequence. The Cretaceous sequence begins with the Bluesky/Gething Formation and ends with Wapiti Formation rocks (FIGURE 2.3), although some formations pinch out in this region. Palynology of mudstones, occurring above and below kimberlite intersections, indicate the Shaftesbury, Dunvegan and Second White Specks Formations as host rocks for B H H kimberlites, as well as rocks from the Smoky Group (TABLE 2.1; Ashton Mining of Canada Inc., 2000). Kimberlites from the Buffalo Head Hills have been dated using U/Pb from perovskite between 83.2 ± 6.68 M a and 87.6 ± 4.6 M a and one date at 64.5 ± 1 . 6 Ma.(Ashton Mining of Canada, Inc. Internal Report on Work, 2000). 9 Table 2.1: Stratigraphy and palynology of mudstones adjacent to kimberlite in the Buffalo Head Hills Region (Ashton Mining of Canada, 2000). Formation Age Environment Shaftesbury Latest Albian - early Cenomanian Lower estuarine Upper Shaftesbury Early Cenomanian Upper estuarine Upper Shaftesbury-Dunvegan Early Cenomanian Upper estuarine Upper Shaftesbury-Dunvegan Early Cenomanian Estuarine bay/lagoon Dunvegan Early Cenomanian Upper estuarine Dunvegan Second White Specks Cenomanian Mid-Turonian Estuarine bay/lagoon Marine anoxic Smoky Group Late Turonian-early Senonian Open marine 10 The palynological evidence together with the U/Pb dates, indicate a minimum time span of 98.9- 83.2 M a for kimberlite volcanism ( F I G U R E 2.3). Paleo-environments during this time period ranged from marine, as evidenced by shale units in the Shaftesbury formation (Dufresne et al. 2001, Bhattacharya and Walker 1991) to non-marine, recorded in the Peace River and Dunvegan Formations (Dufresne et al. 2001, Bhattacharya and Walker 1991). As well, there are intervals where the environment is entirely unconstrained for periods up to 4 to 8 Ma , which may coincide with kimberlite volcanism (Dufresne et al. 2001). As a result the paleo-environment during kimberlite volcanism is unconstrained. However, sedimentary rocks adjacent to kimberlite consist of shale, mudstone and minor sandstone, which implies a marine to marginal marine or tidal flat environment where relief would be minimal. The bedding angles observed in the underlying mudstones and sandstones are flat lying, and the rocks do not indicate a paleo-slope. The character of the rocks imply that host sediments were likely wet and unlithified at the time of kimberlite eruption. 2.2.4 Quaternary Outcrop in this area is sparse as the region is covered in Pleistocene glacial deposits. The deposits range in thickness up to several hundred metres (Pawlowicz and Fenton, 1995). T i l l overlying the kimberlites in this study is clay-rich and unconsolidated. 11 2.3 Classification of rocks as kimberlite Kimberlites are commonly referred to as hybrid rocks (Mitchell, 1995; Scott Smith, 1992; Clement, 1982) which consist of material originating from mantle-derived xenoliths, crustal-derived xenoliths, and primary phases crystallizing from the kimberlite magma (Clement, 1982; Mitchell, 1995). Disaggregation of xenoliths within the kimberlite magma results in the addition of a wide variety of xenocrysts to the magma (Mitchell, 1995). Many of the xenocrysts can be easily identified, however olivine and phlogopite xenocrysts may be indistinguishable from those phases that have crystallized from the magma as primary phases (Mitchell, 1995). Due to this difficulty in distinguishing xenocrystic from primary material, classification by traditional modal and textural criteria are not suitable for kimberlite (Mitchell, 1995; Mitchell and Bergman, 1991; Scott Smith, 1992). Kimberlites therefore are typically defined mineralogically with the presence of both cognate and xenocrystic phases being considered in the definition. Clement (1982) defines kimberlite as follows: Kimberlite is a volatile-rich, potasic, ultrabasic igneous rock which contains abundant olivine and generally has a distinctive inequigranular texture resulting from the presence of macrocrysts set in a fine-grained matrix. This matrix contains as prominent primary phenocrystal and/or groundmass constituents, olivine and several of the following minerals: phlogopite, calcite, diopside, serpentine, monticellite, apatite, spinels, perovskite and ilmenite. Other primary minerals may be present in accessory amounts. The macrocrysts include olivine, phlogopite, picroilmenite, magnesian garnet, chromian diopside and enstatite. Olivine is extremely abundant relative to the other macrocrysts all of which are not necessarily present. In addition to macrocrysts smaller grains belonging to the same suite of minerals occur. Kimberlite may contain diamond but only as a very rare constituent. Kimberlite commonly contains 12 inclusions of mantle-derived ultramafic rocks. Variable quantities of crustal xenoliths and xenocrysts may also be present. Kimberlite is often altered by deuteric processes mainly involving serpentinization and carbonitization. Other definitions have been provided for kimberlite (e.g. Mitchell, 1995) however this definition is generally accepted with the exception that diopside is not considered a primary groundmass mineral (Mitchell, 1995; Scott Smith, 1992). This mineralogical classification is typically applied to fresh hypabyssal rock where crystallization was sufficiently slow to allow the development of typomorphic mineral assemblages (Mitchell, 1995). Classification of rocks from the Buffalo Head Hills region is problematic because the rocks are volcaniclastic and the rapidly quenched juvenile pyroclasts may not have crystallized diagnostic groundmass minerals. In addition they are intensely altered so that the original mineralogy is no longer present. As a result definitive classification of B H H rocks is not undertaken herein. The mineralogy of B H H rocks includes serpentine, carbonate, olivine macrocrysts, olivine phenocrysts, groundmass phlogopite, phlogopite phenocrysts, spinel and perovskite as well as a xenocryst assemblage indicative of kimberlite (Hood and McCandless, 2004). The minerals observed along with the presence of a classic kimberlitic indicator mineral suite (Hood and McCandless, 2004; Carlson et al., 1999) supports the classification of Buffalo Hills pipes as probably kimberlites. Supporting this conclusion is the work of Carlson et al. (1999) where these rocks are classified as Group 1 kimberlite based on the analysis of microphenocrystal atoll-textured spinel and microphenocrystal mica. The spinel exhibit magmatic trend 1, the compositional trend associated with group 1 kimberlite (Mitchell, 1995). The mica exhibit compositional evolution from phlogopite to aluminous phlogopite, which is consistent with the kimberlite trend of Mitchell (1995). The microphenocrysts were selected from the interior of juvenile pyroclasts and as a result are considered representative of the kimberlite magma at the time of eruption. 13 Chapter 3: Volcanic Facies 3.11ntroduction A volcanological model for B H H kimberlites requires knowledge of the distribution and geometry of the bodies and component deposits. Because these bodies rarely outcrop at the present day surface, magnetic, gravity and electromagnetic geophysical methods are used to infer the surface dimensions of these bodies. Seismic data have been used to model their subsurface geometry, however these data are limited to a few bodies. Typically kimberlites are located using geophysical methods and then drilled to determine whether or not a kimberlite is present. If kimberlitic material is intersected, a new kimberlite body is inferred. Rarely is further volcanological work done to test whether the body represents a new kimberlite, a distal deposit from a previously discovered centre or reworked material. In volcanic provinces with multiple centres, it is common for volcanic deposits to be widespread and for extra-crater deposits from different centres to overlap. Dri l l core is the main source of information for interpreting these bodies, however the available core is limited by two facts. Firstly, most of the bodies, although diamondiferous, are sub-economic and only a few holes are available from a single body. Secondly much of the drill core was completely destroyed in the processing for diamond grade estimates. As a result, the data set for any one body in the B H H field is limited. Nonetheless, careful examination of the available material indicates the probability that these rocks represent specific deposits types and that an idealized distribution of volcanic facies can be constructed. These observations are presented in this chapter and are hereafter referred to using their numeric designations, K6,K11, K252, K281, K296 and K300. These kimberlites are located within the Ashton joint venture Buffalo Head Hills claim block (FIGURE 2.2). For bodies K6 , K l 1 and K281, a single drill core per body was available for study. Three drillholes were 14 available from each of K296 and K300, and a total of 5 drillholes were available for study from body K252. 3.1.1 Deposit Characteristics Volcanological assessment of the kimberlite bodies focuses on deposit geometry, clast populations and depositional units. These characteristics are used to investigate processes associated with volcanic fragmentation, transportation, deposition and alteration as well as the effects of subsequent reworking, burial and late alteration. Distinguishing these processes forms the basis for an idealized model for kimberlites in the. B H H . Deposit geometry is a reflection of pre-deposifional relief on the depositional surface, volume of the material deposited, explosivity or fragmentation during eruption, transportation processes and post-depositional erosion (Cas and Wright, 1987; Fisher and Schmincke, 1984, Dellino et al., 2001). Observed volcanic geometries worldwide range from Hawaiian shield volcanoes, which are as much as 200km across, to small negative relief maars only 100's of metres across (Cas and Wright, 1987; Lorenz, 1986). Different geometries are also associated with different clast populations, deposit dispersals and deposit characteristics, which provide evidence for eruptive mechanisms and styles. Pumice clasts produced from Plinian eruptions differ substantially in morphology and distribution from spatter produced through Hawaiian style fire fountaining (Cas and Wright, 1987, Fisher and Schmincke, 1984). Deposit types also vary, with specific deposit types commonly being associated with specific volcanic styles (e.g. base surge deposits and phreatomagmatism). As a result the clast populations, geometries, and deposit types of B H H kimberlites are investigated, as these aspects bear directly on understanding the eruptive processes of these bodies. 15 Once a facies model for BHH kimberlites is established, the distribution and geometry of deposits will be compared to models proposed for kimberlites worldwide, which include steep sided vertical pipes (South Africa), as well as tuff rings and extra-crater deposits (Saskatchewan) (FIGURE 3.1). The inferred geometry of BHH kimberlites is further compared to a broad spectrum of non-kimberlite volcanic geometries in order to determine affinities with other types of volcanism. Kimberlite eruptions have not taken place in recorded time and as a result comparison of these bodies with younger, better-understood volcanoes will serve to clarify the aspects where kimberlites are analogous to other magma types and those aspects where they are unique. 3.2 Volcanological Classification of Kimberlites Historically kimberlites have been classified according to the scheme proposed by Clement and Skinner ( 1985) , which was modified by Field and Scott-Smith ( 1999) . This classification separates kimberlite into three main zones, the crater zone, diatreme zone and root zone, based on textural and structural observations (FIGURE 3.2). This classification is useful for discriminating magmatic rocks (hypabyssal facies kimberlite) from fragmented magmatic rocks that comprise diatreme facies kimberlite and crater facies kimberlite. At the present level of exploration (~200m depth), all of the BHH kimberlite deposits are volcaniclastic and would represent the crater facies with respect to the Field and Scott Smith ( 1 9 9 9 ) model. As a result, this study is primarily concerned with further facies analysis of volcanic rocks. Key steps in this process are the discrimination of pyroclastic rocks from reworked pyroclastic rocks as well as identifying the different types of pyroclastic and reworked deposits. Critical to this analysis is comparison with well-studied volcanic deposits, a summary of which is presented in P L A T E 1. 16 B Kimberlite Pipe Variation (schematic) southern Africa Lac de Gras Saskatchewan 1 km present tratioit surface I ' w o i l j s t i i kimberlite Reset! i men ted kimberlite Didtreme kimberlite kimberlite Figure 3.1:Variations in cross-sectional geometry of kimberlite in A.Canada (modified from Field and Scott Smith, 1999) and B. Globally (modified from Kjaarsgard, 2004). 17 JUVENILE MAGMA TEXTURE Magmatic Magmaclastic Descriptive Terminology r 1 C lasts Unifomi Segregationary ( TeKtu^ed T e n u r e d Globular Segregationary Inter-Clast Matrix I 1 [interpretation!" f . Examples flow-banded matrix-supported t-RVK 'monotonous dast-supported b e d d e d STRUCTURE Descriptive Terminology Cognate Olivine Content >tS% Macrocrystic <15%>1% Sparsely Maaocrystic <1% Phenocrystic Magmaclast / Olivine Size Classification Very coarse Coarse Medium Fine Very fine Very very-fine vc >\Q mm c S-10 mm M 2-5 mm F 0.5-2 mm VF 0.2-0.5 mm WF <0.2 mm Xenoliths / Autoliths Content Type >15% •4rrai Breccia f ~ Crustal Uthic-j L Mantle Autolimic -4mm Microbreoda Exotic Xenocrystic Heterolilhic Textural classification of kimberlites Pipe Zone Textural interpretation Facies CZ Crater Zone + VK Volcanidastic kimberlite = Crater-facies + PK Pyroclastic kimberlite = Crater-fades •4- RVK Epiclastic kimberlite = Crater-facies DZ Diatreme Zone TK Tuffisitic kimberlite breccia = = Diatreme- facies RZ Root Zone + HK Hypabyssal kimberlite = = Hypabyssal-facies Pipe zone and facies terminology of kimberlites Figure 3.2: Kimberlite classification scheme of Field and Scott-Smith (1999). 18 Some of the terminology used herein differs from that proposed by Field and Scott Smith (1999). The term Pyroclastic Lapilli in the Field and Scott-Smith (1999) classification is replaced by the term Pyroclast herein as the term lapilli has size implications. According to Cas and Wright (1987), a pyroclast is "any fragment released in a volcanic explosion or eruption" and that pyroclasts are separated into three size divisions: blocks and bombs (>64mm); lapilli (2-64mm), and ash (<2mm). The three main types of pyroclasts are juvenile fragments, crystals, and lithic fragments. Agglutinated pyroclasts and accretionary pyroclasts are important secondary types. The identification of pyroclasts is the primary basis for discrimination of pyroclastic rocks from magmatic or fragmented magmatic rocks. 3.3 Overview of kimberlites 3.3.1 Paleo- environment Available geochronology and palynology indicates that B H H kimberlites were likely emplaced contemporaneously with host Cretaceous shale, mudstone, fine sandstone and siltstone (see Section 1.2) in a marine to marginal marine environment characterized by low relief. U/Pb ages, from perovskite, ranging from 83.2 to 87.6 M a indicate that these kimberlites are Cretaceous in age and therefore coincident with the clastic sedimentation at that time. As a result the presence of thick kimberlite units (up to >200 m), together with the paleoenvironment, argues that the thickness of kimberlite deposits is not a function of paleo-topography but largely reflects original volcanic topography, with variable post depositional erosional modification. 19 3.3.2 Kimberlite Geometry Carlson et al (1999) describe the geometry of one body in the B H H field, K14, as having a "flared asymmetrical bowl-shaped structure with a central neck and marginal pyroclastic apron deposits of limited thickness". They further state that seismic profiles of other kimberlites in the province imply similar near-surface flaring. At depth, seismic reflection data indicate that the kimberlites are steep sided as evidenced by near vertical reflectors. Seismic profiles over a few of the kimberlites suggest that they are ~40-150 m across at or near basement contact at depths of ~1600 m below present surface (Carlson et al. 1999). In spite of the limited available drill-hole distribution for each kimberlite, some constraints can be placed on the horizontal dimensions of each body from surrounding oil and gas wells (QC Data Ltd., 1999). None of the wells record kimberlitic material of any significant thickness, although it is possible that small amounts of kimberlitic material could have been missed, because it was not the object of drilling. Therefore it can be assumed that substantial accumulations of kimberlitic material are spatially restricted to areas between oil and gas wells, and represent either the primary volcanic products or near source volcaniclastic products from a kimberlite centre. 3.3.3 Clast Population Nomenclature Before presenting physical descriptions of the kimberlite bodies studied herein, it is important to establish nomenclature such that this information can be easily compared with other studies. Of critical importance is the clast population within the kimberlite. These include juvenile pyroclasts, accretionary pyroclasts, olivine macrocrysts, olivine phenocrysts, country rock fragments and mantle xenoliths. 20 Juvenile and accretionary pyroclasts are perhaps the most useful clasts in evaluating volcanic facies and the criteria used to discriminate accretionary pyroclasts, juvenile pyroclasts and autoclasts are presented in Table 3.1. Autoclasts, while present, are rare in B H H kimberlites. The term juvenile pyroclast is used herein to describe fragments of the molten parent silicate magma that were generated by explosive volcanism. This term does not include autoclastic fragments, which formed by the fracturing of previously solidified magma. The term juvenile pyroclast does include pyroclasts that consist of a mantle of magma nucleated by a clast of a different composition. Other names used for juvenile pyroclast in the kimberlitic literature are: juvenile magmaclast (Field and Scott Smith 1999), and juvenile lapilli (Clement 1982). Juvenile pyroclasts are formed by the explosive disruption of magma and are common in pyroclastic deposits, however they can also occur in reworked volcanic deposits showing varying amounts of abrasion. Rounded and abraded pyroclasts are distinguished from primary spherical pyroclasts because clast boundaries cut both matrix and phenocrysts in the abraded fragments, whereas clast boundaries do not cut phenocrysts in juvenile pyroclasts. The term accretionary pyroclast is used herein to describe clasts that have formed by the accretion of ash-sized material about a nucleus to form concentric rims. The nucleus can either be a different fragment or simply accreted ash. Evidence of abrasion for accretionary pyroclasts include any surfaces that cross cut individual rims. It can be difficult to distinguish altered juvenile pyroclasts from altered accretionary pyroclasts. Accretionary pyroclasts are composed of fine kimberlite ash whereas juvenile pyroclasts are composed of coherent kimberlite magma. Where fresh these two types are easily distinguishable however, alteration can render the two pyroclasts virtually indistinguishable. The term olivine macrocryst is used herein to refer to olivine grains that are indeterminate in origin, but many are likely xenocrystic. The term xenocryst is not used to 21 Table 3.1: Comparison of the distinguishing features of juvenile pyroclasts, accretionary pyroclasts and autoliths. Adapted from Cas and Wright (1987), Fischer and Schminke (1984) and Mitchell (1995). Juvenile Accretionary Autoclast Formation Fragmentation of melt Continuous growth through Fragmentation of previously fragment collision and solidified kimberlite.May be accretion in a dilute mixture a previous magmatic phase, of gas vapor and ash. Rims Contacts between rims are sharp. Multiple rims occur but are rare. Mineralogy and texture can vary between rims. Contacts between rims are gradational to diffuse and rarely sharp.Similar mineralogy in each laminae, although grainsize variation is possible. Homogeneous. No rims expected. Shape Fluidal, cuvilinear to spherical in shape Spherical to sub-spherical shapes. Can have knobbly surfaces. Shapes are blocky to angular from fragmentation although rounding can result from abrasion or milling. Clast Clast boundaries do not cross-Boundaries cut constituent crystals. Texture Texture is inherited from the magma; in the case of kimberlite, the magma typically has an inequigranular texture. Quench textures in fine groundmass may be observed in fresh examples. Clast boundaries do not cross-cute constituent crystals, however broken crystals may be present. Textures should be relatively homogeneous, however grainsize variation between rims could occur. There is a tendency towards finer grains due to the difficulty inherent in accreting large grains.Where fresh, individual ash particles should be observed. Clast boundaries cut both matrix and constituent crystals. Texture is inherited from the solidified rock In the case of kimberlite, an inequigranular texture would be expected. Nucleat ion Nucleated and non-nucleated types occur. Nucleated and non-nucleated types occur. Non-nucleated Minera logy Mineralogy not necessarily Mineralogy is typically consitent Mineralogy not necessarily consistent with with surrounding material. consistent with surrounding material. surrounding material. 22 describe these grains as it is commonly not possibly to determine whether an individual grain formed as a result of crystallization from kimberlite magma or whether they represent mantle samples. Furthermore, the term macrocryst is used widely in kimberlite literature to describe these grains and as a result is used for consistency. Individual crystals commonly have rounded edges, and possess abundant internal fractures. Evidence for abrasion of these grains includes broken or truncated grains, however due to their possible xenocrystic origin, many of these grains could have been broken long before deposition. The term olivine phenocryst is used to describe the generally small <2 mm, euhedral, generally non-fractured olivine grains which are interpreted as having formed as a direct product of the kimberlite melt. Evidence for abrasion of these grains would include rounded edges and truncated/broken grains. Leahy (1997) used abrasion of olivine phenocrysts to discriminate primary tephra fall from reworked strata in the Fort a la Corne kimberlites in Saskatchewan. Country rock and mantle xenoliths include fragments of all rock types intersected during kimberlite ascent. Shale, mudstone, sandstone, carbonate rock, granulite facies metamorphic rock and peridotitic xenoliths are present as accessory clasts. Evidence for abrasion of these fragments is not considered as all are abraded to some extent, as would be expected during entrainment, volcanism, and sedimentation. Shale, mudstone and sandstone fragments also occur as accidental clasts incorporated by pyroclastic flows or surges. Interclast matrix varies within and between the bodies. The nature and composition of the matrix provides evidence for proximity to the vent and processes of deposition. Three types of matrix are identified. 1. Colloform carbonate occurring as interclast matrix, is open space fill. In many samples open space still remains at the centre of the carbonate fil l . This type of fill is interpreted as post-depositional fill from late fluids that may or may not be related to kimberlite emplacement. 23 2. Translucent isotropic serpentine matrix is enigmatic. Clement (1982) describes primary serpentine in kimberlite as occurring with a range of properties, but in particular it is noted that it can occur as colourless, structureless, isotropic serpophite that occurs in pools. This description is based on serpentine from hypabyssal or magmatic rocks, however it is very similar to the serpentine observed as interclast matrix in B H H kimberlites. Webb et al. (2003) interprets this material in pyroclastic rocks of the Victor pipe as syn-depositional precipitation of a serpentine 'glass' phase, as deposits containing this fill are commonly matrix supported. However, another possibility is that this material may represent hydrothermal replacement of fine ash with an associated volume gain that creates a pseudo-matrix supported texture. Serpentine is the most common alteration mineral observed in B H H kimberlites, and as a result the possibility that this material is an alteration product cannot be excluded. 3. Cryptocrystalline carbonate and serpentine (± clay minerals) is common as interclast fill and could represent alteration of fine ash, precipitation from magmatic fluids, or cementing of clast supported units through alteration. 4. Fine clay minerals and clastic material derived from surrounding sediments also occurs as interclast fill but only in deposits of reworked kimberlite mixed with surrounding sediments. 24 3.3.4 Data Presentation Data for this thesis was collected f rom macroscopic observations o f dril lcore and microscopic observations f rom polished thin sections. T h e data obtained f rom examination o f dril lcore is presented in stratigraphic sections constructed from each dril lhole examined. W i t h i n the stratigraphic sections the main features recorded are average grainsize, m a x i m u m sizes for each clast type present and modal percentage o f each clast type present relative to the rock volume. Features such as bedding, cross-bedding, or massive structure are indicated by pattern fills, wh ich are explained in Plate 2. Included in the stratigraphic sections are descriptive names for each interval, an interval reference letter, as wel l as columns indicating the lithofacies association o f each interval and the interpreted process o f deposition. T h e data obtained through examination o f polished thin sections is presented mainly in the form o f photomicrographs that are discussed in the text. 25 3.4 Discussion of Bodies 3.4.1 BodyK6 A total of 13 diamond drillholes, and 6 reverse circulation drillholes have been put into the geophysical anomaly representing body K 6 (FIGURES 3.3 and 3.4). The surface expression of this body is interpreted as 16.6 hectares in size. Two vents are interpreted for this body (Ashton Mining of Canada, pers. comm.) one in the vicinity of hole K6-1, and another in the vicinity of hole K6-13, the hole examined in this study. The features of body K 6 are summarized in T A B L E 3.4. Geometry Drill-hole K6-13 intersected 200.9 m of kimberlite unconformably overlain by 50.3 m of glacial till (FIGURE 3.4,3.5). Although little information regarding the morphology of Body K 6 could be derived from the drill-hole examined, nearby drill-holes (FIGURE 3.4) bottom in mudstone at shallower depths, after intersecting intervals of kimberlite. This indicates that the drill-hole examined is likely to represent kimberlite deposited within a crater or vent (TABLE 3.4). Lithofacies Two distinct lithofacies are present within this body (FIGURE 3.5). Lithofacies one (LI , FIGURE 3.6A) composes the bulk of material present within the hole. Lithofacies 2 (L2, FIGURE 3.6B) is restricted to the interval between 102.7 and 126.8m where it occurs interstratified with lithofacies 1 in metre long intervals. Finer beds of L2 show distinct grain alignment whereas coarser beds of L I are relatively massive and featureless. Lithofacies 1 and 2 are considered coarser and finer equivalents, and therefore their characteristics (clast population etc.) are 26 584500 mE 584750 mE 585000 mE 585250 mE 585500 mE 585750 mE 584750 mE 585000 mE 585250 mE 585500 mE 585750 mE Figure 3.3: Outlines of bodies K6 and 252 as defined by gravity contours. The darker black lines overlaid on body 6 are surface expressions inferred from magnetic data. Drillhole locations for each body are indicated by the diamond within a circle.The drillholes examined in this study are in bold. 27 Figure 3.4: Sections showing distribution of kimberlite, host sedimentary rocks and overlying glacial sediments in drillholes through body K 6 . Drillholes are projected to line of section. A. Northern part of body 6 . B. Southern part of body K 6 . Drillhole 6-13 was studied in this thesis. 28 Table 3.4: Characteristics of kimberlite deposits in body K6 Clast Lithofacies Geometry Population Matrix/ Intergranular fill Structures Alteration Interpretation Deposit Type Lla and Lib : Medium-grained olivine crystal-rich tuff Vertically persistent over more than 200m 3P - A Type la: AP Microcrystalline - R carbonate and OM - A serpentine OP - A Type lb: Granular CX - A to colloform carbonate MX - R JP - R AP - N/O OM Type 2: - R Microcrystalline OP - A carbonate and serpentine; CX - P Weak grain orientation, massive featureless Rock is overall fresh however alteration is more intense with depth Proximal fall-back into crater; Rapid cementation in hot environment seals permeability and results in preservation of deposit Pyroclastic fall L2: Fine-grained olivine crystal-rich tuff Occurs as metre thick intervals over a 20m intersection Distinctly bedded with grain alignment Intensely altered, but with rare fresh olivine centres Fluctuation in wind direction or intermittent surges produce sudden variation in grainsize within these beds Pyroclastic fall, Pyroclastic surge MX - N/O Clast Types: JP= juvenile pyroclast, AP= accretionary pyroclast, OM = olivine macrocryst, OP = olivine phenocryst, CX = country rock xenolith, MX = mantle xenolith Clast abundance: A=abundant, P=present, R=rare, N/0= not observed 29 M E A N GRAIN SIZE M a x i m u m ash lapilli . r . , , H v l C F M C Clast Size (mm) | | j | | | 0 2 5 10 20 50 > Percentage Oast Type JP A P LC OM/OP 0 Rock Type IR Lithofacies Interp ' , ' o -Z D ZTTJ <4 <1 <1 N / O <1 <1 <1 <1 <1 N / O N / O N / q <i 60 60 65 75 65 70 ^ 5 65 65 70 70 1 <1 <1 <1 <1 <1 N / O N / O N / O <1 N / O <1 <1 <1 N / O <1 N / O <1 N / O N / O <1 N / O N / O Clay rich till Massive Olivine Crystal-Rich Kimberlite with a mixed Serpentine and Carbonate Matrix - frequent >1cm olivine crystals a n d peridotite xenoliths - juvenile pyroclast observed with e m b a y e d b o t t o m - deposit is poorly sorted and massive in appearance ..... Interlayered coarse olivine crystal-rich kimberlite and fine-grained olivine crystal-rich kimberlite. Estimate corresponds to a fine layer Estimate corresponds to a coarse| layer Estimate of fine layer Massive Olivine Crystal-Rich Kimberlite with a mixed Serpentine and Carbonate Matrix - Dark mixed carbonate a n d serpentine +/-chlorite matrix Massive Olivine Crystal-Rich Kimberlite with a Carbonate Matrix - High angle b e d d i n g contact - Load features beneath large clasts in the upper unit Intensely Altered Massive Olivine Crystal Rich Kimberlite with a Carbonate Matrix - Bedding highlighted by alteration Massive Olivine Crystal-Rich Kimberlite with a Carbonate Matrix LI a L1a/L2 L1a Lib Lib Lib PF PF/PS PF PF PF PF Figure 3.5: Stratigraphic section for vertical hole K6-13. Grain size: F= fine, M= medium, C=Coarse. Clast type: JP= Juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast,OM/OP = olivine macrocryst/olivine phenocryst,0=opaques, P=phlogopite. W = Interval reference.//iferp= Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode ofdeposition: S= normal sedimentation, PF=pyroclastic fall, PS= pyroclastic surge. EOH= end of hole. 30 discussed together. The overall clast populations of the lithofacies are similar except that L2 has fewer large fragments (olivine macrocrysts, juvenile pyroclasts and country rock xenoliths). Grain orientation or weak planar fabric is present and varies from 90-60° to core axis (tea) and defines bedding planes. The clast population is enriched in both olivine macrocrysts and phenocrysts, and depleted in the constituents that form the fine-grained matrix of juvenile pyroclasts. The deposits are moderately sorted with grain sizes ranging from <lmm up to several centimetres. Lithofacies 1(L1)(FIGURE 3.6A) is olivine crystal-rich with an average grain-size of 5mm although clasts vary in size between <1 mm and 2.5 cm. L I has no observable depositional structures other than weak grain orientation. Two subtypes are identified. A mixed serpentine and carbonate matrix distinguishes the first type, which occurs in the upper part of the hole (FIGURE 3.5b,c,d). A carbonate matrix characterizes the second subtype (FIGURE 3.5 f,g). There is a distinct contact between these two subtypes that occurs at 201.1 m depth. This contact is irregular and inclined at ~45° to core axis (tea). Load features interpreted as bomb sags are identified on this surface (FIGURE 3.6 c). Lithofacies 2(L2, FIGURE 3.6B, 3.5c) is olivine crystal-rich with a distinct fabric defined by grain orientation. Average grain-size is on the order of 0.5 mm. Both lithofacies are relatively fresh with limited alteration except for intervals towards the bottom of the hole (201.1, 225.3, 226.2, 241.3,) where alteration is pervasive and texturally destructive. Intervals of intense alteration contain numerous calcite veins. Clast Population Juvenile pyroclasts are not abundant (FIGURE 3.6A,D) but can locally constitute up to 10% of the rock. They are relatively small, ranging in size from <0.5mm up to 2cm with an average size of l-2mm. They range from sub-spherical and nucleated to amoeboid non-nucleated (FIGURE 3.6A, A N D D RESPECTIVELY). Juvenile pyroclasts contain variable amounts of olivine phenocrysts 31 Figure 3.6: Lithofacies of body K6. A. LI is composed of olivine macrocrysts and phenocrysts, with rare peridotite xenoliths, juvenile pyroclasts, and accretionary pyroclasts. The inset shows a juvenile pyroclast with embayed bottom. B. L2 is composed mainly of ash-sized olivine crystals and is finely bedded on a sub-centimetre scale. Dark elongate mudstone xenoliths are oriented parallel to bedding (dashed line). Saw marks run from upper left to lower right. C. Contact between LI a with mixed serpentine carbonate matrix and LI b with carbonate matrix. Note load features beneath the large altered xenolith and beneath the olivine la pi Hi. D. Juvenile pyroclasts occur as non-nucleated amoeboid bodies in a matrix of colliform carbonate. E. Accretionary pyroclast with fine rims. Olivine macrocrysts have fresh centres and altered rims. F. Subhedral olivine phenocrysts and rounded olivine macrocrysts in a matrix of colliform carbonate are typical of both lithofacies (crossed polars). Abbreviations: OM = olivine macrocryst, PX = peridotitk xenolith, JP=juvenile pyroclast, SX = shale xenolith, CX = carbonate xenolith, AP = accretionary pyroclast, OP = olivine phenocryst. 32 and olivine macrocrysts in a very fine-grained mesostasis of predominantly serpentine with lesser amounts of carbonate. Some pyroclasts show indentations on their lower margins, which may suggest that they were still plastic when deposited (FIGURE 3.6A) Accretionary Pyroclasts.are rare and only identified in thin section. These clasts are round in two dimensions and show concentric rims with aligned minerals (FIGURE 3.6E). Rims consist of fine ash-sized fragments commonly altered to fine serpentine and carbonate minerals. The observed clasts are <1.5 cm in size. Olivine Macrocrysts (FIGURE 3.6A,D,E,F) occur throughout, but are most common in coarser-grained units. They vary from fresh to completely pseudomorphed by carbonate minerals and serpentine. They range in size from 5 mm to 2.5 cm, are rounded and fractured, and constitute between 15-25% of the rock. Where serpentinized, fine-grained chromite is common in the alteration assemblage. Olivine Phenocrysts (FIGURE 3.6D,F), in contrast, are rarely fresh, and typically pseudomorphed by fine carbonate minerals and serpentine. They are typically <0.5 mm, but range up to 2 mm. These grains are typically euhedral to subhedral showing little to no evidence of abrasion and compose 35-45% of the rock. Both olivine phenocrysts and olivine macrocrysts typically have alteration rims and many also have thin magmatic selvedges; however it is commonly difficult to distinguish the magmatic selvedges from the alteration rims. Country Rock Clasts make up 5-10% of the rock and include variably altered shale (FIGURE 3.6 B andF), carbonate rock, and hematite/magnetite clasts. These clasts tend to be larger near the top of the hole than at depth. Most xenoliths are subangular and range in size from <1 mm to 7 cm. Carbonate rock xenoliths are white to pink to grey in colour (FIGURE 3.6c). Shale xenoliths are typically smaller and black (FIGURE 3.6B and F), but can be green when altered. Occasionally shale xenoliths are observed that show evidence of soft-sediment deformation. Rare iron-oxide rich xenoliths are bright red or metallic. 33 Mantle Xenoliths (FIGURE 3.6A) are peridotitic and range in size from about 1cm up to 5 cm, and usually have thick reaction rims. Some peridotitic xenoliths contain reddish purple garnet. Single garnet xenocrysts commonly have kelyphitic rims and are 2-3 mm in size. The rare peridotite xenoliths and single garnet xenocrysts are randomly distributed in lithofacies 1 and are absent in lithofacies 2. A single anomalous phlogopite rich kimberlitic clast was identified. The clast mineralogy and the presence of clast boundaries that crosscut internal phenocrysts indicate that this clast is probably an autoclast of a previous generation of kimberlite. Rare blebs of mixed hematite, magnetite, and pyrite are found ranging in size from approximately 2 mm to 2 cm. These constitute <1% of the total rock volume. Late calcite, quartz and bitumen filled veins occur throughout. Structures and Sorting Lithofacies 1 is relatively structureless (FIGURE 3.6A) whereas lithofacies 2 is subtly bedded (FIGURE 3.6B). Both units are enriched in pseudomorphed and fresh olivine crystals relative to estimates of the primary magma from juvenile pyroclasts. Summary and Interpretations The characteristics observed in body six are summarized in T A B L E 3.4 and the proposed geometry and deposits are presented in FIGURE 3.7. Body K 6 is dominated by crystal-rich volcaniclastic rocks which can be generated pyroclastically, epiclastically or by some combination of these two processes. Epiclastic reworking wil l act to sort grains resulting in the separation of different fragments, it wi l l act to abrade, fragile or irregularly shaped pyroclasts which would lead to the physical breakdown and removal of these pyroclasts, and it wil l act to round the corners are edges of crystals. The 34 Figure 3.7: Proposed geometry and deposits of body K6 based on logging of hole 6-13 and utilizing information from holes 6-07 6-08 and 6-10 presence of fresh, non-abraded olivine crystals, non-abraded accretionary and juvenile pyroclasts, is suggestive of pyroclastic rather than epiclastic origin for the rocks observed in body K 6 . Additional support for the interpretation of these deposits as pyroclastic lies in the fact that this intersection of kimberlite is continuous to depths exceeding 200 m; which exceeds the amount of relief expected in this area, and indicates the presence of a vent or crater. Surrounding drill-holes bottom in mudstone at shallower depths, indicating that this location is more proximal to the source of volcanic material. It is considered unlikely that a crater would remain open to depths exceeding 200 m and then subsequently be in-filled through resedimentation. It is more likely that pyroclastic fall and slumping would continuously fill the crater as eruption proceeded. Pyroclastically generated crystal-rich deposits are commonly associated with highly crystallized magmas and further crystal enrichment through sorting (Cas and Wright, 1987). The abundance of single olivine crystals occurring with or without a magmatic selvedge is a common feature of kimberlite deposits (Clement 19.82). This crystal enrichment could be attributed an overall high abundance of dissolved H 2 0 and C 0 2 in the kimberlite melt that exsolves and accordingly increases the proportion of olivine crystals with respect to melt, and would result in deposits that are enriched in olivine crystals relative to juvenile magmatic fragments. Another mechanism for concentrating olivine crystals is sorting of material within the vent, a volcanic plume or column, pyroclastic flows and pyroclastic surges (Cas and Wright, 1987). Accretionary pyroclasts are direct evidence of a volcanic column, or cloud, where fines were circulating. The massive nature of L I (TABLE 3.4, FIGURE 3.7) is consistent with proximal fall back from a pyroclastic column. As a result, the crystal-rich nature of body K 6 deposits was likely enhanced through sorting and elutriation of fines in a volcanic column. The intermittent fine beds in L2 likely represent a distinct change in the depositional regime, which could either represent intermittent surge events or slight changes in column 36 dynamics, such as wind direction. The lack of alteration implies that the deposits were quickly cemented and as a result rendered impermeable to fluids that would otherwise result in intense carbonate and serpentine alteration. It also suggests that this unit was involved in only one major event and not affected by later fluids. Rapid cementation may be more probable in a near vent hot environment rather than in distal or reworked deposits. Several features observed in this deposit bear directly on fragmentation processes. The first feature is dominantly non-vesicular, fluidal, juvenile pyroclasts. Fluidal juvenile pyroclasts can be associated with either phreatomagmatic or exsolution driven eruptions. In phreatomagmatic eruptions, these fragments would be considered passive fragments that form as a result of acceleration of the magma due to the expansion of steam; however, these fragments are not diagnostic to the process of explosive interaction of magma and water (Lorenz et al. 1991). Magmatically fragmented juvenile clasts usually reflect their explosive origin by virtue of a high vesicle content, unless the magma is so fluidal that the fragment flows plastically on landing (Cas and Wright 1987). The shapes of juvenile pyroclasts observed in body K6 , can occasionally exhibit embayment features that indicate some degree of flow upon landing (FIGURE3.6A). In exsolution driven eruptions, non-vesicular juvenile pyroclasts could represent dry magma clots, or previously vesiculated pyroclasts where vesicles have not been preserved. The lack of vesiculation in pyroclasts can be explained through several different mechanisms, including extensive degassing resulting in dry magma clots, resorption of volatiles after deposition into minerals such as calcite and dolomite, pervasive micocrystallization that overprints pre-existing vesicles, the low-pressure collapse of vesicles and intense alteration that obscures vesicles. A second feature that bears on eruptive process is the presence of accretionary pyroclasts. Accretionary pyroclasts are often associated with phreatomagmatism (Cas and Wright, 1987). Accretionary pyroclasts require fine ash generated as a result of fragmentation 37 during a violent eruption. The generation of abundant fine ash from low viscosity mafic magmas is typically associated with phreatomagmatism where magma interacts with water or wet sediments, although ash is also generated from purely magmatic eruptions (Parfitt, 1998); Formation models for juvenile pyroclasts also involve the presence of abundant water, such as would be present in phreatomagmatic eruptions (Gilbert and Lane, 1994). Although the exact paleo-environment at the time of kimberlite eruption is unconstrained some inferences may be made. The first is that this kimberlite would have been intruding wet, poorly lithified sediments (Section 2.2.3). White (1996) argues that phreatomagmatism is actually more likely between impure coolants (wet sediments) and rising magmas due to better mixing between melt and wet sediment. As a result this environment is suitable for the occurrence of phreatomagmatism either subaqueously or subaerially. The presence of accretionary pyroclasts suggests subaerial eruption however it is noted that these clasts can form subaqueously if special conditions are invoked (Cas, pers. comm.). A lack of well developed sorting may also imply subaerial eruption, as subaqueously deposited material is typically much better sorted (Fisher and Schminke 1984). These deposits are interpreted as subaerial pyroclastic deposits, dominated by pyroclastic fall with minor surge and it is likely that at least part of the eruption occurred subaerially. 38 3.4.2 BodyK11 The 10 drillholes into body K l 1 include 5 diamond drill cores and 5 reverse circulation holes. The outline of body K l 1 as inferred by geophysics is 5.4 hectares (FIGURE 3.8). A summary of the characteristics of body K l 1, described below, is presented in T A B L E 3.5. Geometry Dril l hole DDH11-01 was examined from body K l 1; the remaining drill holes had been processed and no significant intervals of material remained. The top of the kimberlite lies unconformably beneath till (FIGURES 3.8 and 3.9). The base of the hole ends in kimberlite. Surrounding drill holes also contain considerable intersections of kimberlite, although two holes bottom in Cretaceous sedimentary rocks. These two holes enter the sedimentary rocks at 152.3 and 157.6 m. This contact could represent the paleo-surface or the bottom/sides of a crater. Lithofacies Two lithofacies are distinguished in body K l 1 (FIGURE 3.9). Both are massive and essentially unsorted with an average grain size of 2mm although clasts range in size from >0.5-60 mm. Different juvenile pyroclast populations distinguish the two lithofacies as does matrix material and degree of clast alteration. A gradational contact separates the lithofacies between 67 and 77 m depth. Overall the rocks in both units are matrix supported and contain a large amount of broken crystals. Accretionary pyroclasts are not observed. Clast Population Juvenile pyroclasts range in abundance up to 30%. In lithofacies 1 (LI) pyroclasts are rare at top and become more abundant (<10%) with depth. They are concentrated in distinct 39 6-19500 mE 613s56mE f l M M m g 61»S56ine 6 i»7ggl^E~ 8T?755TSE 61SS66mE 619600mE 6l9650mE 619700mE 619750mE 619800ITIE B DDH11-05I 20 18.3m 60 R C 11-04 134.1m 120 R C 11-03 134.1m DDH11-03 150.3m A' f,.uoi5r« ' 6330100^'" Figure 3.8: A. Magnetic expression of K11 showing distribution of drill holes and inferred (bold lines) surface outline of kimberlite. B. East-facing cross sectional view of drill holes showing distribution of kimberlite, host sedimentary rock and overlying quaternary sediment. Hole DDH11-01 was examined in this study. 40 MEAN GRAIN SIZE ash lapilli F M C F M C | | | | | | 0 2 5 10 20 50 > JP AP LC OM/OP 0 Maximum Clast Size (mm) Percentage Clast Type Rock Type IR Lithofacies Interp .•'<i>' V-T-T'." °'~r-tt^;-^.'.v.;:o ;^ •<'.°-:'-'o;,'.°'.' • •.' • •.'•'•'o.'•','••! V ' - ' v ' °' -°' • - V -• Y . b '. - ." o*. '.°y'•'. °;\°'w>\-'. • • o','.'.' ;o',;'. . ; o",- .• ° . '. • o- / ' . ' 0 ; . ' °V , •°'.' • • • °- •'•'o•'"'.'0-° •' • •'.°/V '. .'• °. '•'. • „ . ' • ; • ; V ' - ' o ' - '° ' • • o '. ' ° • ..' ' ' . •'•*.'. .o".' •°" ° •' - ' • • '.o'.' • .' .o- * • '. • ' . ° ' . • ',• .' •"; ' ° ' - ' o ' - ° ' -°' • O', .• O . ', o • •'.'(>'. *0',' . • . -', 'o • 'o,', ' ',• o '. . '. * o", • • • • ' '. . ; • ' . ' 0 . ' • ; . • ; . ° . - V ' • ° . V -d •'• ' • • ° '• • ° - • ' ' ' ', . ; o . ' . ' . 0 ,'• ; , . ; 9'y/^;':':'o;.'py : ', . ; o'. *•'. " e , ' • ;o".;'. . ; • ° . ' . ' o ' . -o •' ' . 0 ' . ' . ' .O . ' . 'Q . ' . 0 " . ' "jy , ' o . v . ' 15 25 60 65 60 15 <1 <1 <1 <1 <1 <1 Clay rich till Massive Crystal-Rich kimberlite with fluidal juvenile pyroclasts -Juvenile pyroclasts have uniform single phase internal textures - Olivine is typically fresh but often has altered periphery The contact between L1 and L2 is gradational around 70m Massive Crystal-Rich kimberlite with ragged multiphase juvenile pyroclasts -Juvenile pyroclasts have complex internal textures -Olivine is intesely altered to serpentine LI L2 PF PF Figure 3.9: Stratigraphic section for vertical hole Kl 1 -01. Grain size: F= fine, M= medium, OCoarse. Clast type: JP= Juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst, 0=opaques, P=phlogopite. IR = Interval reference. /nferp= Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode of deposition: S= normal sedimentation, PF=pyroclastic fall. EOH= end of hole. 41 Table 3.5: Characteristics of kimberlite deposits in body K l l Lithofacies Geometry Clast Population Matrix/ Intergranular Structures Alteration fill Interpretation Deposit Type LI: Massive Crystal-Rich Kimberlite L2: Massive Crystal-Rich Kimberlite Continuous intersection. Lithofacies are in gradational contact. JP - R AP - N/O OM - A OP - A CX - P MX - N/O Translucent, microcrystalline serpentine Massive with subtle grain orientation Olivine is dominantly fresh. Elutriation of fines and lack of abrasion indicate a pyroclastic origin Pyroclastic fall infilling crater or vent JP - R AP - N/O OM - A O P - A CX - P MX - N/O Cryptocrystalline mix of serpentine and carbonate Massive with subtle grain orientation Olivine intensely altered to serpentine. Ragged multiphase pyroclasts. Elutriation of fines and lack of abrasion indicate a pyroclastic origin. Pyroclasts appear to be recycled in more than one volcanic event and olivine crystals are intensely altered. Pyroclastic fall infilling crater or vent Clast Types: JP= juvenile pyroclast, AP= accretionary pyroclast, OM = olivine macrocryst, OP = olivine phenocryst, CX = country rock xenolith, MX = mantle xenolith Clast abundance: A=abundant, P=present, R=rare, N/0= not observed 42 horizons and have predominantly fluidal shapes and less common spherical shapes. Nucleated (FIGURE3.10A) and non-nucleated (FIGURE3. 10B,C) pyroclasts are present. Nuclei can be xenolithic (FIGURE 3.10A,D) or cognate crystals. Pyroclasts have a higher carbonate content than the surrounding inter-clast matrix and have sharp to diffuse margins. The diffuse margins may be a function of alteration. In lithofacies 2, pyroclasts are abundant and constitute up to 30% of the rock. These pyroclasts commonly contain multiple phases (FIGURE 3. 10C,D,E,F) with individual phases distinguished by subtle changes in texture. These pyroclasts can be either nucleated (FIGURE 3.10D,E) or non-nucleated (FIGURE 3.10C,F). Mineralogically, they are similar to pyroclasts in LI, however L2 pyroclasts are either agglutinated or autolithic, as they contain multiple phases. The individual phases have not been examined in detail however it is considered likely that these phases represent a single magma where the differences between phases are largely a function of variable formation conditions for each pyroclast phase (e.g. cooling rate). These clasts may be analogous to composite lapilli described by Fisher and Schmincke (1984). Their boundaries are irregular and could be considered ragged (FIGURE 3.10 E,F) and they commonly contain broken olivine crystals. These pyroclasts also have high carbonate content in the matrix. Olivine macrocrysts are abundant in both units (15-20%) and typically occur as broken fragments and less commonly as whole macrocrysts. These clasts range in size from <0.5 mm up to 4 mm. In LI, olivine is very fresh (FIGURE3.1 1A,B,C,D) with minor serpentine alteration on the periphery or along fractures. This alteration consists of microcrystalline yellow translucent isotropic serpentine. In L2, olivine is extensively altered to zig-zag textured serpentine with little fresh olivine remaining (FIGURE 3.1 1,E,F). Olivine phenocrysts (FIGURE 3.12) constitute 10-15% of both lithofacies. Phenocrysts range from euhedral to sub-rounded (FIGURE 3.12A-D); rounded shapes appear to be principally a 43 Figure 3.10: Juvenile pyroclasts occurring in body Kl 1. A. Juvenile pyroclast nucleated on shale xenolith. B. Non-nucleated juvenile pyroclast containing olivine macrocrysts and phenocrysts. C. Juvenile pyroclast containing smaller pyroclast. D. Juvenile pyroclast with two distinct rims nucleated on intensely altered manlte xenolith. E. Ragged pyroclast with complex internal texture containing broken olivine macrocrysts. F. Pyroclast with complex internal texture. Abbrev/af/ons: JP-juvenile pyroclast, OM-olivine macrocryst, OP-olivine phenocryst, MX-mantle xenolith, SX-shale xenolith, P-pyroclast, either autolithic or agglutinated. 44 Figure 3.11: Olivine in body K11. A. Fresh olivine grains in translucent serpentine matrix. B. Olivine macrocrysts in cryptocrystalline serpentine matrix. C. Broken and fractured olivine crystals in a matrix of translucent serpentine. D. Panel C in crossed polars. E. Intensely serpentinized olivine macrocysts in a mixed carbonate serpentine matrix. F. Panel E in crossed polars. Abbreviations: JP-juvenile pyroclast, OM-olivine macrocryst, OP-olivine phenocryst, MX-mantle xenolith, SX-shale xenolith. 45 0.3 m m Figure 3.12: A. subrounded olivine phenocrysts next to garnet with kelyphite rim. B. Large phlogopite clast surrounded by relatively euhedral olivine phenocrysts. C. Subhedral elongate olivine phenocrysts in a serpentine matrix. D. Shapes of olivine grains are similar both within and outside the magmatic rim. E. Enlargement of magmatic rim shown in D. F. Panel E shown in reflected light. Note the relict euhedral texture of olivine phenocrysts both inside and outside the magmatic rim. Abbreviations: JP-juvenile pyroclast, OM-olivine macrocryst, OP-olivine phenocryst,MX-mantle xenolith, SX-shale xenolith, Gnt-garnet 46 function of concentric serpentine alteration along grain boundaries (FIGURE 3.12E,F). Phenocrysts range from fresh in LI, to completely pseudomorphed in L2. Country rock clasts are typically angular to sub-rounded, range from <lmm up to several em's in size, and are variably altered. These typically make up less than 5% of the units but locally can occur up to 15%. They are commonly mantled by magmatic material. Shale, carbonate, and metamorphic basement xenoliths are observed. Mantle xenoliths are intensely altered and consist of phlogopite, serpentine and garnet. Garnets with kelyphitic rims and garnet fragments occur throughout (FIGURE 3.12A). These are a minor constituent and make up less than 2% of the rock. Lithofacies 1 contains a translucent microcrystalline serpentine matrix (FIGURE 3.11A,B,C,D) which is enigmatic in origin (See section 3.3.3). Lithofacies 2 matrix is a fine mix of serpentine and carbonate (FIGURE 3.1 1E.F), which could be primary magmatic, late fill, or the alteration of fine ash. Structures and Sorting The entire kimberlitic intersection in hole K l 1-01 is generally massive with no distinct bedding, contacts, layers or other internal structure except for a consistent but weak grain orientation inclined at 50° to the vertical core axis. Although this orientation is subtle and not always observable, where seen, it did not change orientation throughout the sequence. The concentration of country rock xenoliths tends to increase with depth. Overall the deposit is poorly sorted, but enriched in olivine crystals relative to estimates from juvenile pyroclasts. Summary and Interpretations Body K l 1 is a massive poorly sorted crystal-rich deposit of considerable thickness ( T A B L E 3.5) that extends to depths exceeding 150.3 m. The depth of continuity and drillcore 47 intersections suggest the presence of a crater, and that the deposits are pyroclastic or at least proximal volcaniclastic deposits (FIGURE 3.13). Crystal enrichment could be the result of volcanic or epiclastic processes however the lack of significant abrasion on olivine crystals, the preservation of delicate kelyphite rims on garnet, the lack of significant amounts of country rock xenoliths, and the presence of juvenile pyroclasts all argue against resedimentation, suggesting a pyroclastic origin. A further consideration is that the olivine population is relatively fresh in LI. The non-abraded juvenile pyroclasts and fresh olivine crystals argue against resedimentation. Alteration is more intense in L2, however the multiphase pyroclasts and lack of significant abrasion argue more for recycling of pyroclasts in a volcanic regime than resedimentation. These lines of evidence suggest that these deposits represent primary pyroclastic products. As with body K6, there are a number of processes that can form massive unstructured deposits (PLATE 1), including pyroclastic flows, debris flows, proximal pyroclastic fall and surge, and slumping. The crystal-enrichment of the deposit indicates elutriation of fines, which can occur in pyroclastic fall, surge or flow. Elutriation of fines in a pyroclastic column and subsequent deposition through proximal pyroclastic fall is considered the best alternative to explain these deposits although proximal surge or flow deposits may be very similar in appearance. If these deposits settled through water this could also account for the lack of fines however this would still result from pyroclastic fall, simply through a different medium. There is a distinct gradation with depth to higher carbonate content in the matrix, more intensely altered olivines and ragged, multiphase juvenile pyroclasts that typify L2. This could reflect several processes, two of which are the following: 1. The environment within which the clasts are being deposited becomes increasingly reactive with depth and/or residence time, or 2. The material encountered at depth has been involved in more eruptive events and recycled more times. The second possibility would explain the multiphase pyroclasts observed at depth within 48 11-03 R C 1 1 - ° 4 RC 11-03 RC 11-01 25m 25m Figure 3.13: Proposed geometry and deposits of body Kl 1 based on logging of hole 11-01 and utilizing information from holes 11 -02, 11 -03,11 -04, RC11 -01, RC11 -03, RC11 -04,and RC11 -05. 49 the hole. One possible explanation for increased recycling of pyroclastic material with depth is that these would represent older deposits when eruptions had a lower magmatic content and were more phreatic or volatile driven than in later eruptions. Presumably, this reflects a larger magmatic component in the later eruption products, which could correspond to an increase in magma flux or height within the conduit. Fisher and Schmincke (1984) describe similar pyroclasts that they call composite lapilli. they interpret these clasts as forming when lava droplets are ejected into steam above the level of a magma column, are subsequently quenched and fall back to acquire a new rind of lava. This interpretation is consistent with the recycled material being related to eruptions with a lower magma to fluid ratio, where pyroclasts would be re-entrained. This process could result from either phreatomagmatism or exsolution driven magmatism. Both alternatives are more probable in an intra-crater environment than in an extra-crater volcanic sequence. Heat would likely increase with depth in a crater. Furthermore fragments deposited within a crater would be available for recycling in repeated eruptive events. Minimal alteration indicating quick cementation would also be more likely in an intra-crater environment where volatiles are streaming through a pyroclastic pile. As with body K6 the paleo-environment of eruption of K l 1 is unconstrained and the deposits have been considered above in a subaerial context. However, the processes discussed could occur in a subaqueous environment as well. Water is a more effective sorting medium than air and the crystal concentration observed could result from aqueous elutriation of fines and density sorting of clasts. The absence of accretionary pyroclasts would also be expected in a subaqueous environment. There is however no unequivocal evidence for either subaerial or subaqueous eruption and deposition. 50 3.4.3 BodyK252 Body K252 is located approximately 300 metres from body K6. Although K252 is described as a distinct body herein, the close spatial relationship with body K6 suggests there could be a genetic relationship. This potential relationship was important to bear in mind as the body was examined. The 1.7-hectare surface expression of body K252 has been inferred using geophysics (FIGURE 3.3). Of the 12 diamond drillholes into body K252 (FIGURE 3.14), five were available for study (K252-01, K252-03, K252-08, K252-09, K252-12). Some holes have substantial intervals of kimberlitic material; other holes pierce multiple interlayers of kimberlite and mudstone (FIGURES 3.14-3.19). The main features of body K252 are presented in table 3.6. Geometry Body K252 is morphologically the most complex of all the bodies examined. Few of the drillholes contain extensive thicknesses of kimberlite. Instead most drillholes contain interstratified units of volcaniclastic kimberlite and mudstone (FIGURES 3.-15-3.19) indicating that deposits are separated in time. Some beds can be correlated between holes whereas others cannot (FIGURE 3.20). Lithofacies Eight kimberlitic horizons are distinguished as distinct lithofacies. There are, in addition, two kimberlite-clast bearing shale breccia units. Clast types, structure, matrix and other features distinguish lithofacies, however in many instances the variability of these features within an interval is greater than the variability between intervals. As a result, several of the lithofacies are distinguished by stratigraphic correlation, and the character of the lithofacies changes laterally 51 " t=«mE "k-VmE DDH252-04 OS 7m * s ' * * £ DDH252-01 108.2m ^. D D H 2 5 2 - 0 9 153.3m 160 DDH2S2-11 168.5m 25 SO metres 1SD to fcO teo 100 120 DDH252-10 135m D D Kimberlite Overburden Sandstone/ Mudstone ODH252-12 213.1m r i v . i u E Mi' ' ; inE Figure 3.14: North-facing cross section showing drillholes from body K252. Drillholes are projected to line of section.The drillholes indicated in bold are included in this study. See figure 3.3 for plan view. 52 M E A N GRAIN SIZE ash lapilli F MC F MC | | | | 0 2 5 10 20 50 > JP A P LC OM/OP 0 M a x i m u m Clast Size (mm) Percentage Clast Type Rock Type IR Lithofacies Interp , r i °l?y, EOH IM/O |M/0 N/C1N/q<1 >15 >15 >5 >5 50 >i >1 N/Q N/dN/q N/dN/q N /q Clay rich till contains a few rounded kimberlite pebbles at base Olivine crystal kimberlite rich Mudstone Breccia Contains kimberlite clasts Intensely altered accretionary pyroclast -rich kimberlite contains relict accretionary pyroclasts, relict olivine and country rock xenoliths Mudstone L1 L2 RVK " P F " PF Figure 3.15: Stratigraphic section fo hole K252-01 inclined at 70'. Grain size: F= fine, M= medium, C=Coarse. Clast type: JP= Juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst, O=opaques, P=phlogopite. IR = Interval reference. Interp= Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode of deposition: S= normal sedimentation, PF=pyroclastic fall, RVK= reworked volcaniclastic kimberlite. E O H - end of hole. 53 0.5 1 2 5 1020 M E A N GRAIN SIZE ash lapilli F M C F M C 1,1 I I I 1 o M a x i m u m Clast Size (mm) 2 5 10 20 50 > JP Percentage Clast Type AP LC OM/OP O Rock Type IR Lithofacies Interp y.:&: • ' i ' . ' - rT' •o-'<i>'.0'-rT' ': '.''.6'-— °'-~r ^ • = ; < i > ' . ° ' - ^ : ' • ?o"":' O / V'n- •.' • •'.' o • © 0 0 . . . o • ® O 0 J). „ • © • a). „ - .© ' . ° ° • ® . • • . a . o • (S) ° .; ® . ' o °®v.'. ® : -o ; " ® 'o'.'o EOH <s2><1 15 <1 <1 1 25 10 35 80 30 40 <1 <1 <1 <1 <1 <1 <1 <1 Clay rich till Olivine crystal kimberlite rich Fine Bed Coarse Bed Normal grading Intensely altered accretionary pyroclast -rich kimberlite Mudstone Breccia Coherent mudstone with intervals of mudstone breccia and kimberltic component. Bedding angles 55-65' tea. Intensely altered accretionary pyroclast -rich kimberlite Mudstone with kimberlitic component Massive mudstone with minor disruption. Minor kimberlitic component. Massive kimberlite Mudstone Massive kimberlite Mudstone Mudstone L1 L2 L3 L4 L4 RVK PF/RVK S/VK PF/RVK VK R V K Figure 3.16: Stratigraphic section for vertical hole K252-03. Grain size: F= fine, M= medium, C=Coarse. Clast type: JP= Juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst,0=opaques, P=phlogopite. IR = Interval reference. Anferp=Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode ofdeposition: S= normal sedimentation, PF=pyroclastic fall, RVK= reworked volcaniclastic kimberlite. VK= volcanicalstic kimberlite. EOH= end of hole. 54 MEAN GRAIN SIZE ash lapilli F MC F MC LLLLU Maximum Clast Size (mm) Percentage Clast Type 0 2 5 10 20 50 > JP AP LC OM/OP 0 Rock Type IR Lithofacies Interp O • o x •; • Q . . , • • V p b v - ° ' x ' . ' ° ' . . . •. • o • • • ' • •>.•.•.• o •. • - 4 ) ' / ' ' o • ' ^ O / o -. • o • x x . • • . y ! . • • > • „ • • • • • . • . x ;o.. »c>-°.r •,.y fo f**. • '• '.'•".*: ,-\ .; .;py\'q >>;";.'-'-;:"',P' "„*•?*''* ,x !•*'>. ••".'o * •.•\ •'•o , l •*/?• VV\ .o* • > * ' t v % i - ' ' lb ' '•':**. V'.'/'d ••*».•> .•.ff'.Oo'-,''.-?; EOH Clay rich till Intensely altered, olivine crystal - rich kimberlite Intensely altered accretionary pyroclast -rich kimberlite Intensely altered, olivine crystal - rich kimberlite Densely packed olivine crystals intensely altered and often irresolvable from matrix material! White ~2mm olivine grains give the rock a speckled appearance however many of the olivine grains are altered to the same colour as matrix material. L I L2 L1 RVK PF/RVK RVK Figure 3.17: Stratigraphic section for hole K252-08 inclined at 60'. Grain size: F= fine, M= medium, C=Coarse. Clast type: JP= juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst,0=opaques, P=phlogopite. IR = Interval reference. Interp= Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode of deposition: S= normal sedimentation, PF=pyroclastic fall, RVK=Reworked volcaniclastic kimberlite, EOH= end of hole. 55 MEAN GRAIN SIZE M a x i m u m nv^FNu: ClastSizelmm) | | | | | | 0 2 5 10 20 50 > Percentage Clast Type JP AP LC O M / O P 0 P Rock Type IR Lithofacies Interp - o • - : ; ' i - . v — j 3 ; 7 7 . ; r T ; . V ; . o •o-'cv °'-rr' .° °'-r-ro • ° . - j . ' o r r 0 ^ •o • • • • •, -\-^:^:°'+.°°:—\ r-o °-"T"- r EOH 10 10 65 65 40 65 >1 >1 >1 >1 >1 >1 >1 >1 Clay rich till Mudstone - Beds flat lying at around 85' tea - Beds are 5mm-1 cm thick Mudstone is fragmented at the lower contact and kimberlite occurs between fragments. Bedded kimberlite iGrain supported, normally gradeel Mudstone Brecciated above the kimberlite Bedded kimberlite Densley packed, well sorted. JMQBljaiadiQa Mudstone brecciated at the top Bedded kimberlite Mudstone Bedded kimberlite Mudstone . i t L3 L4 L4 L4 VK VK S " F T V K " Figure 3.18: Stratigraphic section for vertical hole K252-09. Grain size: F= fine, M= medium, C=Coarse. Clast type: JP= juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst, 0=opaques, P=phlogopite. IR = Interval reference. Interp= Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode of deposition: S= normal sedimentation, PF=pyroclastic fall, VK=volcaniclastic kimberlite. EOH= end of hole. 56 MEAN GRAIN SIZE ash lapilli F M C F M C | | j | | | 0 2 5 10 20 50 > JP AP LC OM/OP 0 Maximum Clast Size (mm) Percentage Oast Type Rock Type IR Lithofacies Interp . ' •S^^ 'Vo- - ' 1 . " ^ . - ' - . * - ' " . ' an" //0> -III' EOH / \ >1 >1 <1 10 45 50 >1 30 50 10 5 5 35 15 15 10 45 15 10 10 35 15 40 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 Clay rich till Mudstone Bedded Kimberlite - Normally graded - Bimodal fragment population - Clast supported Mudstone Top of mudstone is disturbed Mudstone and Kimberlitq Mudstone Mixed Mudstone and Kimberlite Mudstone - contact sharp but undulose - mud draped and disturbed Fine-Bedded Kimberlite coarse bed fine bed Mudstone Fine-Bedded Kimberlite Mudstone Oil sand m L4 L5 L6 L7 L8 PF S ' R V K " RVK PS/RVK S W Figure 3.19: Stratigraphic section for hole K252-12 inclined at 70'. Grain size: F= fine, M= medium, OCoarse. Clast type: JP= juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst,0=opaques, P=phlogopite. IR = Interval reference. /nferp=Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode of deposition: S= normal sedimentation, Pf=pyroclastic fall, RVK=reworked volcaniclastic kimberlite PS= pyroclastic surge. EOH= end of hole. 57 (FIGURE 3.20). Stratigraphic correlation assumes a lack of topography and assumes continuity of deposits, which are reasonable assumptions based on the nature of the mudstone country rock and volcaniclastic nature of the deposits. Lithofacies 1 (LI) is a well-sorted olivine crystal-rich kimberlite occurring in intersections K252-01b (FIGURE 3.15), K252-03b (FIGURE 3.16), and K252-08 b and d (FIGURE 3.17). This lithofacies is distinguished by pronounced concentration of olivine crystals, lack of juvenile and accretionary pyroclasts, and excellent sorting (FIGURE 3.21 A-D). This unit is the uppermost intersection of kimberlite in holes K252-01 and K252-03, and is the dominant material in hole K252-08. In all instances it is in unconformable contact beneath overlying till and is in sharp contact with underlying material. Lithofacies 2 (L2) consists of intensely altered accretionary pyroclast-rich kimberlite occurring in intervals K252-01d (FIGURE 3.15), K252-03c and e, (FIGURE 3.16) and K252-08c (FIGURE 3.17). L2 is distinguished by stratigraphic location, intense alteration and close packing of accretionary pyroclasts (FIGURE 3.21E-H). The abundant closely packed accretionary pyroclasts typically have concentric rims of fine ash (FIGURE 3.21H) that is now altered to microcrystalline carbonate and serpentine. This lithofacies is similar to the accretionary pyroclast-rich member of lithofacies 4 (L4) but is separated from L4 stratigraphically and exhibits intense alteration, which commonly masks texture. This lithofacies attains a maximum thickness in hole K252-01 and thins in both holes K252-03 and K252-08. Lithofacies 3 (L3) ranges from olivine crystal-rich to accretionary pyroclast-rich kimberlite and occurs in intervals K252-03e (FIGURE 3.16) and K252-09c (FIGURE 3.18). It is defined based on stratigraphic position, as the materials observed in the two intersections that constitute L3 are dissimilar. In K252-03e (FIGURE 3.16) L3 contains abundant accretionary pyroclasts and is similar in appearance to L2. In K252-09c (FIGURE 3.18), L3 is olivine crystal-rich (FIGURE 3.22A) and similar in appearance to the olivine crystal-rich rocks in L4 (FIGURE 3. 58 Figure 3.20: Proposed geometry and deposits of body K252 based on logging of holes 252-01,252-03,252-08,252-09 and 252-12. 59 Figure 3.21: Lithofacies one A. Fine-grained, olivine crystal-rich, finely bedded kimberlite. B. Medium to coarse grained, olivine macrocryst-rich, bedded kimberlite. C. Intensely altered,fine to medium-grained kimberlite. D. Olivine crystal-rich fine grained bedded kimberlite. Lithofacies two E. Intensely altered accretionary pyroclast-rich kimberlite. F. Detail of nucleated accretionary pyroclast in intensely altered kimberlite. G . Densely packed intensely altered, accretionary pyroclasts. H . Individual accretionary pyroclast Sub-spherical in three dimensions, cross-sectional break surface shown. This pyroclast is finely laminated and nucleated on an altered mantle xenolith. Abbreviations: OM=olivine macrocryst, AP=accretionary pyroclast. White arrows indicate upward stratigraphic facing direction; black dotted lines indicate approximate bedding angle. 60 25209 106.8m 2.5 cm & • I 25212 1 3 2 . 9 ^ ^ 1.5 cm Figure 3.22: Lithofacies 3 A. Resedimented pyroclastic kimberlite. Grainsize increases from right to left,or uphole, indicating that this section is normally graded.Clasts consist predominantly of closely packed olivines and lesser amounts of mudstone clasts. Lithofacies 4 B. Olivine crystal-rich member of L4. Olivine crystal size increases with depth resulting in a normally graded sequence. This rock is well sorted and olivine crystal-rich. C. Olivine crystal-rich member.This sample is dominated by olivine crystals but contains accretionary pyroclasts. Bedding is defined by grain orientation and lies at 35' to core axis. D. Accretionary pyroclast-rich member of L4. Shown is split core containing a large accretionary pyroclast nucleated on an intensely altered mantle xenolith. E. Accretionary pyroclast-rich member of L4. Cross section of core showing abundant sub-spherical accretionary pyroclasts with abundant smaller ~5mm olivine grains set in a fine-grained matrix. This sample has a bimodal grainsize distribution with the accretionary pyroclasts composing one size fraction and olivine crystals dominating the second size fraction. Many of the accretionary pyroclasts display fine concentric rims. Abbreviations: OM=olivine macrocryst, AP=accretionary pyroclast. White arrows indicate upward stratigraphic facing direction, black dotted lines indicate approximate bedding angle. 61 22B,C). Although different in appearance and composition these intervals correlate stratigraphically and are inferred to represent different facies of a single deposit. Lithofacies 4 (L4) is compositionally similar to L3 and ranges from accretionary pyroclast-rich to olivine crystal-rich kimberlite with the two occurring either as gradational to each other or as distinct members. (FIGURE 3.22B-D) This unit encompasses a continuous section of kimberlite in K252-12c (FIGURE 3.19) and several intervals of kimberlite interstratified with mudstone, or mudstone breccia in intersections K252-09e,g, and i (FIGURE 3.18) and K252-03g, i and k (FIGURE 3.16). The intervals likely represent separate depositional events; however, the lack of mudstone intervals in K252-12 may indicate that this is the parent deposit for reworked material in K252-09 and K252-03, or that mudstone intervals pinch out in this area, possibly due to thicker accumulations of pyroclastic kimberlite. This lithofacies is distinguished stratigraphically as there are a variety of textures and structures present. The same sequence of rocks is observed in all three holes, although accretionary pyroclast-rich material is dominant in holes K252-12 and K252-03 whereas hole K252-09 is dominated by olivine crystal-rich kimberlite. Lithofacies 5, 6, 7 and 8 are observed in hole K252-12. These lithofacies are defined stratigraphically and they contain many of the same constituents as previous lithofacies. L5 is olivine crystal rich kimberlite with a significant crustal xenolith component. Accretionary pyroclasts are present but not abundant. Lithofacies 6 consists of olivine crystal rich kimberlite with accretionary pyroclasts present but not abundant. This interval is well bedded and sorted. Lithofacies 7 (L7) is a medium-bedded accretionary pyroclast and mudstone fragment-rich kimberlite. This deposit has cm-scale beds. Accretionary pyroclasts are typically in 5-7mm in diameter. The lack of a bimodal grainsize distribution distinguishes this lithofacies from other accretionary pyroclast bearing lithofacies. In most lithofacies, the accretionary pyroclasts are much larger than other clasts, however in L7 accretionary pyroclasts are in the same size range 62 as other clasts. L8 is an olivine crystal-rich kimberlite that is closely packed and moderately sorted. Bedding is defined by grain size variation. This unit is similar to L3 and olivine crystal-rich material in L4, but differs in its marked lack of accretionary and juvenile pyroclasts. Mudstone breccia lithofacies Mixed kimberlite and mudstone fragment breccias occur in intervals K252-01c, K252-03d, K252-09 g (FIGURES 3-15,3-16 A N D 3-18 RESPECTIVELY). Individual kimberlite fragments are small, l-2cm in size, and consist of angular fragments of previous kimberlite deposits. Detailed clast population descriptions are not completed on the individual fragments due to their small size and rarity. Clast population The distribution of juvenile pyroclasts is variable both within and between lithofacies but they rarely constitute more than 5% of the material present (FIGURES3.15 to 3.19). LI contains almost no juvenile pyroclasts while they are common in L4. There are two types of juvenile pyroclast identified in body K252. Type one contains abundant mica and olivine phenocrysts (FIGURE 3.23 A,B,D,E), and type two is dominated by olivine phenocrysts with minor mica (FIGURE 3.23 A,C,E). Type one is the dominant type observed. Both types of juvenile pyroclast tend to have fluidal shapes, neither type is observed to contain vesicles. They range from >1 mm up to about 1 cm in size. The juvenile pyroclasts observed are typically not abraded. Accretionary pyroclasts are observed in all kimberlite lithofacies excepting lithofacies one and eight. Abundance of these clasts ranges up to >50% of the rock. Accretionary pyroclasts are either nucleated on older pre-existing clasts or non-nucleated, and range in size from about 2 mm up to several centimetres in size (FIGURE 3.21E,H; 3.22B-D). Many of the accretionary pyroclasts have concentric rims (FIGURE 3.24A,B) that are defined by subtle changes 63 Figure 3.23: Juvenile pyroclasts of body K252. A. (Left) Olivine phenocryst-rich juvenile pyroclast with fluidal shape. (Right) smaller phlogopite-rich juvenile pyroclast. B. Fluidal juvenile pyroclast containing phlogopite and olivine crystals (pseudomorph). C.Olivine crystal-rich juvenile pyroclast, and discrete olivine macrocrysts and phenocrysts set in a matrix consisting mainly of microcrystalline carbonate. D. Phlogopite and olivine crystal-rich juvenile pyroclast with fluidal shape occurring in microcrystalline carbonate matrix. Shale xenoliths are common in this interval. E.Two juvenile pyroclasts and abundant discrete olivine crystals occurring in a matrix of cryptocrystalline carbonate.The juvenile pyroclast on the left contains phlogopite and the one on the right does not. Abbreviations: ]P= juvenile pyroclasts, OM=olivine macrocryst, Ph= phlogopite,OP= olivine phenocryst, SX=shale xenolith. 64 in texture or grain size; however, massive types are also common (Figure 3.24c). In three dimensions, these clasts are spherical (FIGURE 3.21H). Similar to juvenile pyroclasts, some accretionary pyroclasts have been abraded as shown by clast boundaries that crosscut internal elements. (FIGURE 3.24D). In most of the lithofacies, however, these clasts are not abraded. Olivine macrocrysts occur in all kimberlitic lithofacies in varying quantities. In L I , they compose between 5 and 10% of the rock, whereas in parts of L4 they constitute up to 60%. Olivine macrocrysts range in size from around 2 mm up to 1.5 cm. They are elongate and sub-rounded, and intensely altered to carbonate and varying amounts of serpentine. Olivine macrocrysts typically occur as discrete grains but can also be included in juvenile or accretionary pyroclasts. The abundance of olivine macrocrysts in a given interval most likely reflects the sorting processes, as they are most abundant in coarser horizons. Olivine phenocrysts occur in all kimberlitic lithofacies, but are most abundant in the fine-grained intervals. Olivine phenocrysts are the dominant clast type in L I , where they constitute ~40-55% of the rock. In contrast, in L3 and parts of L4 they constitute <20% of the rock. They range from subhedral to euhedral in shape and are nearly completely altered to serpentine and carbonate. Olivine phenocrysts occur both as discrete grains and within juvenile and accretionary pyroclasts. Intergranular fill in all lithofacies is dominated by cryptocrystalline carbonate with lesser amounts of serpentine and other clay minerals. This matrix has a dark mottled appearance. Country rock xenoliths are rare in L I and consist of small (~1 mm) subangular shale xenoliths. Country rock xenoliths are most common in accretionary pyroclast-rich lithofacies and tend to be less abundant in lithofacies dominated by olivine crystals. In some intervals (FIGURE 3.22D), country rock xenolith sizes parallel the size of accretionary pyroclasts. The dominant country rock types are mudstone and carbonate rock. Angular fragments are common as are 65 Figure 3.24: Accretionary pyroclasts of body K252. A. Detail of a 4cm in diametre accretionary pyroclast nucleated on an olivine macrocryst.The interior of this clast is unstructured however fine rims are visible towards the periphery (lower left corner). B. Unstructured accretionary pyroclast nucleated on olivine macrocryst. C. Detail of the edge of an accretionary pyroclast showing at least three rims (outlined in white) D. Accretionary pyroclast nucleated on olivine macrocryst with outer edges that cross-cut constituent olivine phenocrysts. Ab6revfaf/ons:AP=accretionary pyroclast, OM=olivine macrocryst, OP=olivine phenocryst 66 fragments showing soft-sediment deformation or embayment by adjacent pyroclasts. Country rock xenolith content typically ranges up to 20%. Structures and Sorting Normally graded beds are pronounced in LA, both in the olivine crystal-rich and accretionary pyroclast-rich horizons. In olivine crystal-rich horizons, normal grading includes all component clasts. In accretionary-rich horizons, there are several areas where normal grading is only observed in a subpopulation of the component clasts. An example of this is in interval K252-12c (FIGURE 3-19) where accretionary pyroclasts and country rock xenoliths increase in size with depth but the other clasts (mainly olivine macrocrysts) maintain a uniform size throughout. All lithofacies are bedded except L2, which is massive in appearance. Bed thickness varies from thin <1 cm in LI, up to ~5 cm in parts of LA. Individual beds are defined by grainsize and clast alignment. Overall, olivine crystal-rich lithofacies have a tendency to be better sorted than accretionary pyroclast-rich facies. LI has excellent sorting. Accretionary pyroclast-rich rocks commonly display bimodal grain populations, which is particularly notable in K252-12 (FIGURE 3-19). Summary and Interpretations A summary of the features described above is given in T A B L E 3.6 and the modeled geometry is presented in FIGURE 3.20. Lithofacies 8 through 5 are interpreted as material deposited as single beds on relatively flat surfaces. Each lithofacies represents a distinct event that interrupted normal clastic sedimentation, represented by shale intervals. L5 through 8 are well bedded and contain varying amounts of olivine crystals and accretionary pyroclasts. The deposits represent varying amounts of resedimentation of primary pyroclastic kimberlite deposit. L7 contains the highest proportion 67 Table 3.6: Characteristics of kimberlite deposits in body K252 Lithofacies Geometry p J^fion Matrix/ Intergranular fill Structures Alteration Interpretation Deposit type LI: Well-sorted olivine (altered) crystal rich kimberlite Laterally continuous, final phase of kimberlite material JP - N/O AP - N/O OM - P OP - A CX - R MX - N/O Cryptocrystalline mix of serpentine and carbonate Very well sorted, distinctive bedding, fluctuating grain size Excellent sorting and absence of juvenile and Intense accretionary pyroclasts suggest sedimentary re-working. Reworked pyroclastic kimberlite L2: Intensely altered accretionary pyroclast rich kimberlite laterally continuous but pinches out JP - R A P - A OM - P OP - P CX - P MX - N/O Clay minerals Massive, Consists of closely packed accretionary pyroclasts Intense alteration that is often textu rally destructive Preservation of delicate clasts suggests pyroclastic deposit Pyroclastic fall L3: Olivine crystal-rich/ Accretionary o s i t i o n a l l y pyroclast rich Laterally continuous but kimberlite JP - P AP - A OM - A OP - A CX- P MX - P Cryptocrystalline mix of serpentine and carbonate Ranges from massive to bedded with normal grading Grades laterally from accretionary Intense pyroclast rich to crystal rich implies facies change Lateral facies change from pyroclastic fall to reworked pyroclastic kimberlite. Clast Types: JP= juvenile pyroclast, AP= accretionary pyroclast, OM = olivine macrocryst, OP = olivine phenocryst, CX = country rock xenolith, MX = mantle xenolith Clast abundance: A=abundant, P=present, R=rare, N/0= not observed 68 Table 3.6 continued Lithofacies Geometry Clast Population Matrix/ Intergranular Structures Alteration Interpretation fill • Deposit type L4: Accretionary pyroclast and olivine crystal-rich kimberlite Changes from a continuous intersection to several intersections seperated by mudstone JP - P AP - A OM - A OP - A CX - P MX - P Cryptocrystalline mix of serpentine and carbonate/clay minerals Bedding, Grading, and a bimodal clast population characterize this lithofacies. Intense Grades both laterally and vertically from accretionary pyroclast rich to crystal rich kimberlite Lateral facies change from pyroclastic fall to reworked pyroclastic kimberlite. L5: Olivine crystal-rich kimberlite with significant crustal component Unconstrained JP - P AP - P OM - A OP - A CX - P MX - N/O N/O Bedding present but subtle Intense Dominantly pyroclastic material. Moderate reworking possible Pyrocastic fall or Reworked material L6: Olivine crystal-rich kimberlite Unconstrained JP - P AP - A OM - A OP - A CX - P MX - P N/O Distinctly bedded, with moderate to good sorting. Intense Dominantly pyroclastic material. Moderate reworking possible Pyrocastic fall or Reworked material Clast Types: JP= juvenile pyroclast, AP= accretionary pyroclast, OM = olivine macrocryst, OP = olivine phenocryst, CX = country rock xenolith, MX = mantle xenolith Clast abundance: A=abundant, P=present, R=rare, N/0= not observed 69 Table 3.6 continued Lithofacies Geometry Clast Population Matrix/ Intergranular fill Structures Alteration Interpretation Deposit type JP - P L7: Medium AP - Accretionary bedded A Bedding is pyroclasts are non-accretionary OM - A defined by abraded. Deposits Pyroclastic surge pyroclast and Unconstrained N/O grain size Intense are well sorted and mudstone OP - A variation. Beds well bedded fragment-rich CX- P are cm-scale suggesting traction kimberlite without abrasion MX - P L8: Olivine crystal-rich kimberlite Unconstrained JP - N/O AP - N/O OM - A OP - A CX - P MX - P N/O Well bedded and sorted Lack of juvenile and accretionary pyroclasts suggests Intense re-working, excellent sorting and bedding indicate traction Current action Clast Types: JP= juvenile pyroclast, AP= accretionary pyroclast, OM = olivine macrocryst, OP = olivine phenocryst, CX = country rock xenolith, MX = mantle xenolith Clast abundance: A=abundant, P=present, R=rare, N/0= not observed 70 of accretionary pyroclasts, of these four lithofacies and may represent an original pyroclastic deposit. In L4, there is a continuous lateral facies change from pyroclastic material to intensely reworked material. In the vicinity of hole K252-12, L4 consists of pyroclastic material. In hole K252-03, L4 is moderately reworked, and in K252-09, L4 is intensely reworked; olivine crystals are concentrated and juvenile and accretionary pyroclasts removed through abrasion and sorting. In K252-12, L4 exhibits very well preserved accretionary pyroclasts and well-developed bimodal sorting, both characteristics of pyroclastic fall deposits. Bimodality in volcanic deposits has been discussed by Sparks et al. (1997) and is commonly the result of the aggregation of ash material that allows it to drop out prematurely. This process is similar to the process by which accretionary pyroclasts are formed. This explanation fits with the types of clasts observed in these deposits that include a high proportion of accretionary pyroclasts. In K252-03, accretionary pyroclasts are abraded and the deposit is crystal enriched indicating reworking. Similarly in K252-09 the material is olivine crystal rich with very few juvenile or accretionary clasts present, which is thought to indicate progressive reworking, sorting and abrasion. L3 records a transition similar to the one described for L4, where accretionary pyroclast-rich primarily pyroclastic material is observed in K252-03 however the stratigraphic equivalent in K252-09 consists predominantly of well-sorted, normally graded, olivine crystals. L3 is structureless in K252-03 but is normally graded and bedded in K252-09. This supports the interpretation of a transition from pyroclastic material to reworked material. In holes K252-03 and K252-01, L2 is interpreted as primary pyroclastic fall composed of closely packed accretionary pyroclasts. The material in K252-08 however may have experienced some degree of re-working as accretionary pyroclasts are not as well preserved and commonly broken. Alteration is intense and as a result many features are obscured, perhaps because of the highly permeable nature of pyroclastic fall deposits. 71 LI lacks juvenile and accretionary pyroclasts, and exhibits excellent sorting. This lithofacies is interpreted as reworked volcaniclastic kimberlite. The absence of pyroclasts supports the interpretation of epiclastic transport as the delicate pyroclasts would be destroyed by abrasion and removed through sorting. Thus this deposit is interpreted as reworked volcaniclastic kimberlite. Lithofacies 1 is finely bedded, indicating a medium to high-energy transport mechanism such as current transport. The mudstone breccias are consistent with brecciation of country rock and previously formed/solidified kimberlite. Mudstone breccias may be analogous to basal lithic breccias that are commonly observed as the first phase in phreatomagmatic volcanism. Mudstone breccia units observed are structureless but uniform and consistent with pyroclastic fall. General Comments The presence of accretionary pyroclasts indicates several things: 1. Transport in a dilute medium such as a pyroclastic column, cloud, surge or flow (Cas and Wright, 1987); 2. A humid environment where condensation of liquid around falling solid particles results in particle accretion (Gilbert and Lane, 1994); 3. The presence of fine ash as an eruptive product, although the size of individual ash grains is indeterminate as the material is altered and primary textures are obliterated; and 4. Lack of significant reworking, Accretionary pyroclasts are quite delicate and breakdown readily upon reworking. As a result non-abraded accretionary pyroclasts are a good, but not definitive, indicator of primary pyroclastic deposits. The large size of many of these clasts and the multiple rims may also indicate continued circulation for extended periods of time. Laboratory experiments and field observations by Ernst (2004) indicate that accretionary pyroclasts form in a similar way to hailstones and that the sizes of accretionary pyroclasts can provide information on eruption column height and variation. 72 Juvenile pyroclasts in body K252 are non-vesicular with fluidal shapes that indicate a very low viscosity magma. The lack of vesicles may be a result of the low viscosity and it is possible that gas phases were not trapped within juvenile pyroclasts but allowed to exsolve at a rate fast enough to preclude the development of vesicles (Clement 1982). There may also be a difficulty in preserving vesicles in melts that do not easily form glass. Rapid quench crystallization may result in the overprinting of pre-existing vesicles. A further consideration is that these fragments formed through shear thinning of an accelerated volume of magma. In this case the mechanism of acceleration could be magmatic exsolution or phreatomagmatism. The characteristics of several of these deposits indicate the involvement of both pyroclastic and sedimentary processes. The pyroclastic deposits identified are likely to be the products of pyroclastic fall or surge as they contain pristine accretionary pyroclasts, can be well sorted and have bimodal clast populations. The reworked deposits are likely to represent current reworking in shallow channels or in a marginal marine environment. 73 3.4.4 BodyK281 Body K281 lies towards the centre of the BHH claim block and is not near to other kimberlites; It has a surface expression of 1.1 hectares inferred from geophysical data (FIGURE 3.25). Seismic modeling over body K281 indicates that it is steep-sided at depths below 200 m. Of the two drillholes through body K281, K281-01 was available for study and the features described from body K281 (TABLE 3.7) are based on this material. Geometry The morphology of body K281 is largely unconstrained as there are only two drill holes in this body, however; the drillhole intersections coupled with the geophysical data suggest that a crater is likely (FIGURE 3.25). If so, the material in drillhole K281-01 represents more vent proximal material than K281-02. This body has been modeled geophysically as being steep-sided at depth and wide in the near surface (Ashton mining of Canada, pers. comm.). Lithofacies Three distinct lithofacies are identified (FIGURES 3.26 and 3.27). Lithofacies 1 (LI) consists of steeply dipping alternating olivine crystal-rich coarser and finer beds (FIGURE 3.27A). Lithofacies 2 (L2) is more massive, coarser grained and contains abundant juvenile and accretionary pyroclasts (FIGURE 3.27B). Lithofacies 3 (L3) contains a bimodal distribution of very coarse (up to 6 cm) accretionary pyroclasts, mantle xenoliths and country rock fragments with a background of finer, 2 mm olivine grains and ~5 mm-1 cm, juvenile and accretionary pyroclasts (FIGURE 3.27 c). There is no stratigraphic order to these lithofacies as they are present at multiple levels in the drill hole. 74 595500 mE 898800 mE 595700 mE S958O0 m E = 1 x7 • — ' 595900 mE 596000 mE 595700 mE 595600 m E 595900 m E 596000 mE B 630l350mN SMttoOmN 20 | Kimberlite Overburden Sandstone/ Mudstone 60 80 DDH281-01 150.3m 6W1?OOrr,N Figure 3.25: A. Magnetic expression of body K281 showing distribution of drill holes and inferred (bold lines) outline of kimberlite from EM data. B. East-facing cross section showing distribution of kimberlite, host sedimentary rock and overlying Quaternary sediment. Hole DDH281-01 was examined in this study. 75 1-MEAN GRAIN SIZE ash la pi II i F MC F MC | | | | | | 0 2 5 10 20 50 > Maximum Clast Size (mm) Percentage Clast Type JP AP LC O M / O P 0 Rock Type IR Lithofacies Interp - ' • ' ° - - o ' ' : 0 O ' - l ' ~P. v .V :° o- • . ' . ]-~--°\— <a\ • ° - , < i ' . ° — ' . ° ' • ° : ' 7 T " 2.' :o-_ • '. • o . O ' - l ' ' . " •T - j . ' o . -V . "P.'.1 o- • . •. ; ' . ~ c • ~ ~ ^ i °0 ~Ti '^ •'o-,i>'.°—'° 2.' . o ^ • '• VCJ'- I ' " • 0 ^ 7 7 : ° o- • • '. ; \-~c '^S 1 . ' •o'c>'. 0 -rr' .° 2' . ' O i l • '• ' • ! 0 ; - H -; .o •-!_' "P.'. ;.° o- • . •. '. "•<• ; ^ i . ° ' . _ i <i-. ' ° o T i ' l v—' • fc:v/?'-^:^-:?::«^:v:;?'-^ ' ® ' i : ? i ^ ' ; ^ . ® ; y ' ' ' ' : £i ej °-.:'-®;-.';:?--'^ ®: '^.-°'.-'<S 10 10 5 10 15 5 10 10 30 5 10 45 50 5 15 10 45 30 55 45 20 55 10 55 30 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 Clay rich till Fine-bedded kimberlite -Alternating layers of coarser and finer grains -Layers oriented at 30'tca. -Bedding <1 cm to 3cm. [Juvenile & Accretionary Pyroclast-Rich kimberlite] Fine-bedded kimberlite fine bed Juvenile & Accretionary Pyroclast-Rich Kimberlite Three types of juvenile pyroclast Massive medium grained kimberlite Accretionary Pyroclast -Rich kimberlite Bimodal grainsize distribution Medium-grained accretionary and juvenile pyroclast - rich kimberlite weakly bedded Fine-bedded kimberlite Massive" med Tu m-"grai neq Jaccretionary and [juvenile pyroclast - rich kimberlite Juvenile & Accretionary Pyroclast-Rich kimberlite! J L1 L3 L1 L3 L2 L3 L2 L I L2 L3 PS PF PS PF PF PF PF PS PF PF Figure 3.26: Stratigraphic section for vertical hole K281 -01. Grain size: F= fine, M= medium, C=Coarse. Clast type: JP= Juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst,0=opaques, P=phlogopite. //?= Interval reference. Interp= Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode of deposition: S= normal sedimentation, PF=pyroclastic fall, PS= pyroclastic surge. E OH- end of hole. 76 Figure 3.27: Body K281. A. End view of core with angled cut resulting in oblong appearance in photo. Lithofacies 1 consists of fine-bedded kimberlite. Fine beds are dominated by olivine crystals and coarser beds contain olivine crystals, juvenile pyroclasts and accretionary pyroclasts. B. Lithofacies 2 (L2) consists of medium-grained, subtly bedded to massive kimberlite. This lithofacies contains a relatively uniform size distribution of olivine crystals, accretionary and juvenile pyroclasts with clasts ranging between 0.5mm and 1 cm in size. C. Lithofacies 3 (L3) is a coarse accretionary pyroclast-rich, juvenile pyroclast and olivine crystal-bearing pyroclastic kimberlite. This rock has a bimodal clast population with large, up to 4cm, accretionary pyroclasts and country rock xenoliths occurring with a second population of typically <5mm olivine crystals and juvenile pyroclasts. Abbreviations: AP= accretionary pyroclast, OM= olivine macrocryst, CX= carbonate xenolith. 77 Table 3.7: Characteristics of kimberlite deposits in body K281 Lithofacies Geometry Clast Population Matrix/ Intergranular fill Structures Alteration Interpretation Deposit type JP - P LI: Fine- AP - R Cryptocrystalline bedded, Occurs as OM - A mix of serpentine olivine crystal- repeated OP-and carbonate-rich intervals A serpentine kimberlite CX - P dominant Alternating coarse and fine beds with cross-bedded and graded beds. Intense MX - N/O Fluctuating energy levels. Minimal abrasion of clasts. Bedding at high angles indicates 'sticky' deposition. Characteristic of wet surge. Pyroclastic surge L2: Juvenile and accretionary pyroclast bearing, olivine crystal-rich kimberlite Vertically continuous; grades with lithofacies three JP - P AP - P/A OM - P OP - P CX - P MX - R Cryptocrystalline mix of serpentine and carbonate -serpentine dominant Transport in dilute Centimetre- particle gas scale Intense mixture. Clasts bedding involved in multiple events (recycled). Pyroclastic fall and possible synvolcanic slumping, JP - P L3: Accretionary pyroclast-rich kimberlite Vertically AP - A Cryptocrystalline Transport in dilute Pyroclastic fall involved in repeated continuous; OM - A mix of serpentine particle gas events (recycling). gradational to lithofacies OP - A and carbonate - Massive serpentine Intense mixture. Clasts involved in multiple Large accretionary pyroclasts indicate two CX -MX -P P dominant events (recycled). prolonged residence in column. Clast Types: JP= juvenile pyroclast, AP= accretionary pyroclast, OM = olivine macrocryst, OP = olivine phenocryst, CX = country rock xenolith, MX = mantle xenolith Clast abundance: A=abundant, P=present, R=rare, N/0= not observed 78 Clast population LI is a crystal rich deposit (FIGURES 3.27A, and 3.28A,) that contains abundant pyroclasts but few definitive juvenile pyroclasts according to the criteria established in T A B L E 3.1. Many clasts occur that could be either juvenile or accretionary pyroclasts, however they are either too small or altered to be distinguished as one or the other (See section 3.3.3) (F I G U R E 3.28B). L2 contains juvenile pyroclasts that are nucleated, non-nucleated, amoeboid and round in two dimensions (F I G U R E 3.28C). Two pyroclast types, olivine phenocryst-rich pyroclasts and phlogopite-rich pyroclasts, are defined based on mineralogy. L3 has the most diverse juvenile pyroclast population of any deposit examined in this study. Several distinct juvenile pyroclast types are common, including olivine phenocryst-rich, phlogopite-rich, calcite lathe-rich and apatite-bearing juvenile pyroclasts. Furthermore juvenile pyroclasts with up to three distinct magmatic rims are present (FIGURE 3.28D). Juvenile pyroclasts range in size from <0.5 mm up to several centimetres, however most are in the 2-5 mm range. Accretionary pyroclasts occur nucleated on olivine crystals (F I G U R E 3.29A) on juvenile pyroclasts (F I G U R E 3.29B), and on country rock fragments. They also occur as non-nucleated bodies (FIGURE 3.29C). The accretionary pyroclasts consist of fine ash along with olivine crystals and fragments. Accretionary pyroclasts can be either massive or finely laminated. Where thin accretionary rims are present (FIGURE 3.29A,B) these are good criteria for distinguishing these clasts from juvenile pyroclasts. In L2 and L3, accretionary pyroclasts are abundant, up to >40%, and large, up to >6cm. In LI accretionary pyroclasts are less abundant (<25%) and generally <2mm in size. Within individual accretionary pyroclasts alternating laminae of coarser and finer material can occur (FIGURE 3.27C). Olivine macrocrysts are abundant in coarse horizons of LI, and throughout L2 and L3. Olivine macrocrysts range in size from ~2 mm up to <1 cm, are typically altered to light green or 79 Figure 3.28: Juvenile pyroclasts of body K281. A. Fine-bedded olivine crystal-rich deposit, containing juvenile pyroclasts, accretionary pyroclasts and mudstone xenoliths. Bedding is defined by grainsize variation. B. Indeterminate pyroclasts (circled in white) that may be juvenile or accretionary in origin, however alteration has obliterated primary textures. C. Two juvenile pyroclasts, the one on the left is non-nucleated, the one on the right is nucleated on a shale xenolith. Both exhibit inequigranular texture. E. Juvenile pyroclast with three distinct magmatic rims.The first rim (1) is aphanitic,the second (2) contains abundant olivine phenocrysts, whereas the third (3) rim contains abundant fine apatite.This pyroclast is broken along the bottom margin. Abbreviations: AP=accretionary pyroclast, JP= juvenile pyroclast, OP= olivine phenocryst, SX= shale xenolith. 80 Figure 3.29: Accretionary pyroclasts from body K281 A. Accretionary pyroclast with many thin rims towards the edge and more massive material towards the centre. This accretionary pyroclast is nucleated on an altered olivine phenocryst. B. Accretionary pyroclast with distinct rims at edge and more massive centre.Centre may be an altered juvenile pyroclast or kimberlitic autolith. Surrounding the accretionary pyroclast are discrete olivine crystals in a matrix of cryptocrystalline carbonate. C. Accretionary pyroclasts with no distinct internal structure. Both nucleated and non-nucleated types occur. Abbreviations: AP= accretionary pyroclast,OP= olivine phenocryst, JP= juvenile pyroclast. 81 black in colour, and composed of serpentine and calcite. Olivine macrocrysts increase in abundance with depth in the coarser units. Olivine phenocrysts are typically in the 0.2-0.5 mm size range although they can be as large as 2 mm. They are euhedral and show no evidence of abrasion, however there are many broken fragments in the very fine beds of LI. Olivine phenocrysts make up to 55% of the material in fine-bedded LI deposits, but can compose as little as 10% of the material in L3 deposits. Country rock fragments are dominantly shale and mudstone derived from the host Cretaceous sedimentary rock sequence. Also relatively abundant are carbonate rock fragments derived from underlying Devonian rocks. Mantle xenoliths, where present, are concentrated in distinct horizons. These xenoliths are typically rounded, composed predominantly of olivine (peridotite), range up to 6 cm in size, and have granular texture. The three distinct types of interclast matrix observed are isotropic translucent serpentine, cryptocrystalline mix of serpentine and carbonate, and granular carbonate. In areas where the isotropic serpentine matrix is present, the units have a higher proportion of matrix. As discussed earlier (section 3.3.3), the implications of this matrix are enigmatic, as it may represent magmatic fluids or later alteration. Granular carbonate matrix is interpreted as late fill. The cryptocrystalline mix of serpentine and carbonate minerals could represent magmatic fluids, replacement of fine ash, or alteration of larger grains that results in a volume gain and pore space fill. Structures and Sorting LI intervals are well bedded generally at high angles, up to 30°to core axis, which correlates to 60° from horizontal. LI intervals consist of laminae ~3 mm to 3 cm thick (FIGURES 3.27A and 3.28A) of slightly coarser and finer olivine grains, pyroclasts and xenoliths. L2 intervals 82 are weakly bedded and can occasionally contain isolated fine beds similar to those in lithofacies one. Overall L2 is massive although wavy and high angle bedding is observed locally (FIGURE 3.27B). L3 lacks obvious bedding (FIGURE 3.27A). LI rocks are very well sorted with respect to size but contain a mix of lithologies and include scattered large fragments up to several centimetres in size. L2 rocks are relatively well sorted with ~lcm accretionary pyroclasts occurring within a matrix of 2-5 mm olivine macrocrysts, <2mm olivine phenocrysts and ~2-5 mm juvenile pyroclasts. L3 rocks are typically unsorted with a bimodal distribution of fragments. Accretionary pyroclasts and country rock xenoliths dominate the coarse size fraction which is dominantly >1 cm in size. The clasts interstitial to these larger clasts are predominantly in the 0.5 -5 mm size range and are dominated by discrete olivine crystals, juvenile pyroclasts, country rock xenoliths and accretionary pyroclasts. Summary and Interpretations Body K281 is continuous to depths below 150 m as a result it is inferred that there is either a crater or large volcanic sequence in this area F I G U R E 3.30. The deposits either represent primary pyroclastic fill of a crater, resedimentation into a crater or accumulation of a pyroclastic pile with resedimentation on the sides. Both juvenile and accretionary pyroclasts are intact and show little evidence of abrasion. Abrasion that is noted could be attributed to pyroclastic transport. Olivine grains also lack evidence of abrasion. It is therefore inferred that these deposits have not been extensively reworked. The multi-rimmed accretionary pyroclasts indicate prolonged transport in a dilute ash vapour medium. Lithofacies 2 and 3 have a massive appearance and occur interstratified with intervals of LI. It is suggested that L2 and 3 represents background near vent pyroclast fall and that LI represents surge events that occur intermittently. 83 Figure: 3.30: Proposed geometry and depostis of body K281 based on logging of hole 281-01 and utilizing information from hole 281-02. 84 LI is finely bedded at relatively high angles (up to 70° in a vertical hole). There are no distinct contacts between lithofacies to suggest that this material represents a rotated block, transitions are typically gradational, and the bedding angles are assumed to be in-situ. Steep bedding angles are encountered near the top of the intersection (40 m) and at depths of 126 m with L2 and L3 intervals occurring in between. The high-angle beds indicate that the depositional plane was close to vertical and fine material was plastered on to the sides of an opening suggesting that the beds represent deposition from near vertical blast events. High angle deposition of thin beds of volcanic material is characteristic of wet surge deposits. LI has characteristics consistent with deposition from a "wet" pyroclastic surge. This interval occurs intermittently indicating surge pulses. Accretionary pyroclasts in L2 and L3 are indicative ash accretion in a dilute turbulent mixture of ash, water vapor and air in a pyroclastic cloud or column. Accretionary fragments formed due to circulation and dropped out near vent due to their large size, whereas olivine crystals dropped out as a result of high density resulting in a bimodal distribution. As a result these deposits are interpreted as pyroclastic fall, with possible synvolcanic slump components as well. The accretionary pyroclasts present are unusual because they are large, up to >6cm, and contain distinct laminae with abundant fragments greater than 0.5 mm in size. Sparks et al. (1997) state that it is very unlikely that fragments greater than 80 pim will be accreted in the formation of juvenile pyroclasts. This indicates that these accretionary pyroclasts are unusual. It is suggested that these deposits represent near vent pyroclast fall, much of which has been involved in more than one event (i.e. recycled). Juvenile pyroclasts with multiple magmatic rims require clast recycling or evolution of the magma with depth. Recycling of material to produce multiple magmatic rims is best achieved 85 proximally where there is a chance for material to fall back into the vent and be re-mantled. This supports the interpretation of these deposits as near vent pyroclastic fall. Considered together, the individual aspects of L1-L3 indicate that body K281 consists of a mix of fine-grained crystal-rich surge deposits (LI) and near vent recycled proximal pyroclastic fall (L2 and L3). The overall relationships between the two types of deposits indicate continuous pyroclastic fall with intermittent surge events that crosscut the fall deposits. Contacts are typically sharp and cross-cutting however in some areas they can be irregular with patches of surge material occurring within the pyroclastic fall close to the contact. The fall material is consistent with repeated involvement in explosive events. The juvenile pyroclast population observed preserves evidence of explosive fragmentation of kimberlite magma. Juvenile pyroclasts are non-vesicular and fluidal to spherical in shape. A lack of vesicles in juvenile pyroclasts is commonly associated with phreatomagmatic fragmentation. However vesicle preservation may be low in melts that do not easily transect the glass transition phase, and as a result pre-existing vesicles may have been obliterated. The fluidal shapes are similar to shapes generated through Hawaiian fire fountaining where shear thinning is the dominant fragmentation method. Despite the uncertainties in formation of juvenile pyroclasts the widespread presence of accretionary pyroclasts requires fragmentation processes with sufficiently high energy to generate fine ash. In low viscosity magmas, like kimberlite, intense fragmentation is typically associated with phreatomagmatic processes, as is the formation process of accretionary pyroclasts. 86 3.4.5 BodyK296 Three drillholes were put into body K296 which has a surface expression of 400mx400m inferred from electromagnetic data (Ashton Mining of Canada 2003b, F I G U R E 3.31). All three holes were available for study and are discussed herein (FIGURES 3.32,3.33,3.34). A summary of the features of body K296 is presented in T A B L E 3.8. Geometry From the inferred size of the body and drill core information it is noted that the deposit thins from more than 190 m thick to 55.5 m thick 237 m away. The geometry indicated by these holes suggests the presence of a crater (Figure 3.31, 3.35) as this exceeds the maximum expected relief in this area at the time of kimberlite eruption (See section 2.2.3). Lithofacies Four lithofacies are identified in body K296. Lithofacies 1 (LI) consists of finely interbedded (FIGURE 3.36) lapilli and coarse ash with well-defined sorting. LI is observed exclusively in hole K296-03. This lithofacies makes up the entirety of kimberlitic material in this hole and is not observed in contact with any other units (FIGURE 3.34). Olivines are buff colour and pyroclasts tend to be dark grey. Lithofacies 2 (L2) is depleted in juvenile pyroclasts relative to LI, and clasts in this lithofacies are almost all broken (FIGURE 3.37). L2 is observed in hole K296-01 and K296-02 (FIGURE 3.32c, 3.33C,D). This unit is difficult to identify as kimberlite due to the high degree of alteration and the absence of relict grain shapes (FIGURE 3.37). Euhedral grains are rare to absent. Lithofacies 3 (L3) consists of carbonate mudstone (FIGURE 3.38A) and finely laminated microcrystalline carbonate rock (FIGURE 3.38B) that contain scattered pyroclasts and relict volcanic grains but are otherwise very fine-grained. They are both composed almost 87 B Figure 3.31: A. Magnetic expression of body K296 showing distribution of drillholes and inferred (bold lines) surface expression of kimberlite from electromagnetic data (Ashton Mining of Canada pers. comm.). B. East-facing cross sectional view of drillholes showing distribution of kimberlite, host sedimentary rocks and overlying Quaternary sediments. All holes shown were examined in this study. 88 M E A N GRAIN SIZE M a x i m u m f^cf^l Clast Size (mm) | I | | | | 0 2 5 10 20 50 > JP A P LC OM/OP O Percentage Oast Type Rock Type IR Lithofacies Interp 5 10 65 60 Clay rich till Clay rich till with minor sand and occasional pebbles. Sandy Mudstone Alternating fine sandy mudstone, calcareous mudstone and unconsolidated sandy clay. Evidence of soft sediment deformation, bioturbation and fossil material as well as water escape structures. Bedding is typically subhorizontal. Fine-bedded kimberlite Fine Bed Coarse Bed L2 RVK Figure 3.32: Stratigraphic section for vertical hole 296-01. Grain size: F= fine, M= medium, OCoarse. Clast type: JP= Juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst, 0=opaques, P=phlogopite. IR = Interval reference. Interp= Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode of deposition: S= normal sedimentation, RVK = reworked volcaniclastic kimberlite. EOH= end of hole. 89 M E A N GRAIN SIZE M a x i m u m ash lapilli . c . , . N v T c F M C Clast Size (mm) | j | | | | 0 2 5 10 20 50 > JP A P LC OM/OP O Percentage Oast Type Rock Type IR Lithofacies Interp ' • o • • V : ' • ; To.'. A . o 7 I * „ *r-p u g "ep;. bY?-"'v~6">% V^-°^ °'<*'-'• b-f-,|S>!--.b :^-^'-?^^V-0i7?!'*V- .b" • V V ° « '•.•:o*,j Vb;- •• •-.•"J |-?-°o'V .Q-"^ -*-''- •-'•^'"cJ'''-^~'";-?-°o'V'r1-.?r??ivV^?*;-.- :.?rj;?:i ^ • • . b • ' • i j ' ^ ' " ' - ' 0 ' o 50 40 10 55 60 10 15 <1 <1 <1 Clay rich till Clay rich till with horizons of sandier material Sandy Mudstone Dark grey containing fine sand. Very fine flat lying beds. M\Td'stone7T^m6erfite Mottled mixed k i m . a n d m u d Intensely altered kimberlite Juvenile pyroclast bearing olivine crystal tuff. Pervasive texturally destructive alteration. Distinct bedding in less altered areas.There are rare normally graded sequences. Dense Carbonate Grades into kimberlitic mudstone! Mudstone Kimberlitic c o m p o n e n t Fine-grained kimberlite Altered but distinctly b e d d e d Massive kimberlite Steady increase in clast size f rom dominantly 1 m m at top to average of 5 m m at 105m below which the grain size stays relatively constant. Bedding is defined by clast size a n d type. Coarser beds contain more juvenile pyroclasts, finer beds contain more olivine crystals. Massive kimberlite with unaltered olivine Olivines in this unit are fresh as o p p o s e d to almost entirely altered above.The intergranular fill is also darker and appears to contain less carbonate. Hypabyssal kimberlite autolith or dyke Crystal-rich horizon within massive kimberlite Massive kimberlite with unaltered olivine Hypabyssal kimberlite autolith or dyke e L2 L3^ L3 L1 L4 L4 L4 L4 RVK RVK PF PF/S PS PF PF PF PF Figure 3.33: Stratigraphic section for vertical hole K296-02. Grain size: F= fine, M= medium, C=Coarse. Clast type: JP= Juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst,0=opaques, P=phlogopite. W= Interval reference. Interp= Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode of deposition: S= normal sedimentation, PF=pyroclastic fall, PS= pyroclastic surge. EOH= end of hole. 90 MEAN GRAIN SIZE M a x i m u m fjjjfcffjlz Clast Size (mm) Percentage Clast Type | I | | | | 0 2 5 10 20 50 > JP AP LC O M / O P O P Rock Type IR Lithofacies Interp 25 — 50 — 75 — 100 — m 0 Clay rich till Sandy Mudstone a S -'•°-„-° — - 0 - . 0 — o • • • • • • ~ - ° ° „-<y °— °° — * - o - 0 -© • . • . ."• • -- ° ' o • • ~ - ° ° = • c ° — ° ° — - ;, — 'Y.K o . ' • ° - - o -° — o— Q — " C "• c Alternating finely bedded calcareous, fossiliferous mudstone and calcareous fossiliferous sandstone. b S B fl1 >w 5 40 <i 0 5 3 70 10 <1 <1 <1 1 Fine-bedded kimberlite Alternating coarse and fine beds. Fine beds consist predominantly of olivine crystals; coarse beds consist of juvenile pyroclasts. Cross-bedding on cm scale fine bed coarse bed c L1 PS/PF 125 — 150 — Sandy Mudstone Fine, black to dark grey/brown bedded to massive mudstone containing interbeds of fine-grained, finely bedded sandstone d S Figure 3.34: Stratigraphic section for vertical hole K296-03. Grain size: F= fine, M= medium, C=Coarse. Clast type: JP= Juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst, O=opaques, P=phlogopite. IR = Interval reference. Interp= Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode of deposition: S= normal sedimentation, PF=pyroclastic fall, PS= pyroclastic surge. EOH= end of hole. 91 Table 3.8: Characteristics of kimberlite deposits in body K296 Lithofacies Geometry Clast Population Matrix/ Intergranular fill Structures Alteration Interpretation Deposit type LI: Fine-bedded, olivine crystal, and juvenile pyroclast-rich kimberlite Thick repetive sequence. Laterally continuous JP - A AP - N/O OM - A OP - A CX - P MX - N/O Colloform carbonate Alternating coarse fine beds, cross-bedding, grading Intensely altered to carbonate rich assemblage Fluctuating energy levels, absence of abrasion Pyroclastic Surge L2: Bedded kimberlite with abundant broken crystals Unconstrained JP - R AP - N/O OM - P OP - P CX - P MX - N/O Mainly carbonate Plane parallel bedding. Alteration can obscure texture Intensely altered to carbonate rich assemblage Broken crystals likely result from sedimentary reworking Resedimented crater fill L3: Fine ash and microcrystalline carbonate Thin horizon JP - P AP - N/O OM - P OP - A CX - P MX - N/O Carbonate Fine plane parallel laminations. Bomb sags on bedding planes Intensely altered to carbonate rich assemblage Volcanogenic ash indicates a period of quiescence Pyroclastic fall L4: Massive olivine crystal and juvenile pyroclast-rich kimberlite Thick vertically continuous interval J P - A AP - N/O OM - A OP - A CX - P MX - N/O Colloform carbonate Mainly massive with some grainsize fluctuations. Intensely altered to carbonate rich assemblage Lack of abrasion, poor sorting and composition indicate a pyroclastic origin Proximal pyroclastic crater fill Clast Types: JP= juvenile pyroclast, AP= accretionary pyroclast, OM = olivine macrocryst, OP = olivine phenocryst, CX = country rock xenolith, MX = mantle xenolith Clast abundance: A=abundant, P=present, R=rare, N/0= not observed 92 Figure 3.35: Proposed geometry and deposits of body K296 based on logging of holes 296-03,296-01 and 296-02. 93 Figure 3.36: Lithofacies 1. A. Cross-stratified beds with individual bedsets containing coarse and fine beds. Coarse beds are dominated by juvenile pyroclasts and fine beds are dominated by discrete olivine crystals. B.Normally graded sequence with coarse juvenile pyroclasts dominant at the bottom and ~1 mm olivine crystals dominant at the top. C. Massive section consisting of fluidal-shaped juvenile pyroclasts. D. Fine bedding. Olivine phenocrysts dominate fine beds,coarser beds consist of larger phenocrysts,and small juvenile pyroclasts.The interclast matrix is composed of colliform carbonate that is interpreted as late open space fill. E. Juvenile pyroclast rich bed. Juvenile pyroclasts have fluidal shapes and occur in a matrix of colliform carbonate that is interpreted as late open space fill.Abbreviations: JP= juvenile pyroclast,OP= olivine phencryst. 94 Figure 3.37: Lithofacies 2. A. Massive, intensely altered, kimberlite with primary textures unresolvable in hand sample. B. Intensely altered material with relict textures which may represent altered olivines. Large phlogopite clast in lower left corner. Both altered clasts and interclast material is carbonate. C. Intensely altered material consisting of discrete altered olivine fragments and rare abraded juvenile pyroclasts with clast boundaries that cross-cut constituent phenocrysts. D.OIivine fragments altered to carbonate within a granular to crystalline carbonate matrix. Abbreviations: Phl= phlogopite, OM= olivine macrocryst JP= juvenile pyroclast. 95 Figure 3.38: Lithofacies 3. A. Well-bedded, volcanogenic mudstone. A distinctive bed that shows juvenile pyroclasts embaying underlying layers is highlighted in white. B. Transition from clastic volcanogenic mudstone at left to finely laminated carbonate at the right. C. Fluidal juvenile pyroclasts.These pyroclasts occur on the bed shown in panel A. Some of the pyroclast boundaries crosscut constituent olivine (altered) phenocrysts (shown by arrows). D. Altered olivine macrocryst also occurring on the bed shown in A. E. Altered olivine crystals occurring aligned in distinct beds. F. Finely laminated carbonate. Abbreviations:VM=volcanogenic mudstone,JP= juvenile pyroclastOM = olivine macrocryst, OP = olivine phenocryst. White arrows indicate upward stratigraphic facing direction. 96 entirely of carbonate. Lithofacies 4 consists of massive pyroclastic kimberlite (FIGURE 3.38A) observed only in hole K296-02 only. This lithofacies contains a similar clast population to LI but it lacks distinctive bedding, grading or cross laminations. The appearance of L4 is variable. It is generally darker at depth and has a greater amount of fresh olivine although discrete altered intervals occur sporadically. Intersections of either thin hypabyssal dykes or autoliths of hypabyssal material occur in several horizons. L4 differs macroscopically from LI in that the pyroclasts are commonly buff coloured and the olivines are dark grey, whereas in LI the reverse is true. In L4, rare accretionary pyroclasts are present (FIGURE 3.39B,C); these are not observed in other lithofacies. The most marked difference between L4 and LI is the lack of bedding of any sort in L4, however variations in clast size do indicate a degree of sorting. Colloform or crystalline carbonate is the dominant interclast fill in all lithofacies except L3 where cryptocrystalline carbonate is the dominant material. Colloform carbonate is the dominant fill especially in coarser beds (FIGURES 3.36,3.37 and 3.39) and commonly is an incomplete fill as there are open pore spaces at the centre of the space between the grains. In L3, the matrix consists of finely laminated carbonate (FIGURE 3.38). These deposits are considerably altered with olivine replaced by carbonate and serpentine. Late alteration consists of the infiltration of bitumen, quartz and carbonate in the forms of veins and intraclast vug fill. In some areas the kimberlite has a mottled appearance as a result of intense alteration, which can be texturally destructive. Clast Population In LI, juvenile pyroclasts.are subrounded to amoeboid in shape and are the only clast type in some beds (FIGURES 3.36c and E), but are intermixed with other clasts in other beds (FIGURE 3.36B AND D). Only one composition of juvenile pyroclast has been identified, suggesting that they are likely to represent a single kimberlite magma. The non-vesicular juvenile pyroclasts 97 Figure 3.39: Lithofacies 4 A. Unsorted, massive pyroclastic kimberlite containing grey juvenile pyroclasts and black olivines in a matrix of white carbonate. B. Altered shale xenolith in upper left is surrounded by fluidal juvenile pyroclasts in a colliform carbonate matrix. C. Juvenile pyroclasts and discrete olivine crystals in a colliform carbonate matrix. D. Discrete olivine crystals and fluidal juvenile pyroclasts in a colliform carbonate matrix. E. Accretionary pyroclast with subtle rims. F. Edge of broken accretionary pyroclast showing fine rims defined by grainsize variation. Abbreviations: JP = juvenile pyroclast, OM= olivine macrocryst, SX = shale xenolith, OP = olivine phenocryst, AP= accretionary pyroclast, Carb = carbonate. 98 are composed of variable proportions of olivine macrocrysts, olivine phenocrysts and fine-grained matrix. In general, olivine macrocrysts and phenocrysts composed about 35% of the juvenile pyroclasts. Juvenile pyroclasts range in size from <1 mm up to 2-3 cm. Grain boundaries are sharp and do not cut across crystal boundaries. The amoeboid to sub-rounded shapes are interpreted as primary. Implicit in this interpretation is that the source magma was very low viscosity and sufficiently hot such that these fragments relaxed to their current shape after fragmentation occurred. In L2, juvenile pyroclasts are generally absent (FIGURE 3.37B,D). Where present they are compositionally and texturally identical to pyroclasts in LI with the exception that grain boundaries are observed to crosscut crystal boundaries (FIGURE 3.37 C ) . This indicates that the pyroclasts have been abraded after cooling. Juvenile pyroclasts are not present in L3, with the exception of one horizon where several juvenile pyroclasts define a bedding surface (FIGURE 3.38 A). These pyroclasts have fluidal outlines interpreted as primary shapes (FIGURE 3.38c), however some of the sides cross cut crystal boundaries indicating that some degree of abrasion is likely. Juvenile pyroclasts in L4 are similar to those in LI (FIGURE 3.39A,B,C,D). In some areas they are buff in colour (FIGURE 3.39A), which contrasts with their dark grey colour in LI. Mineralogically and texturally they are however the same. Accretionary pyroclasts are observed only in LA where they occur up to several centimetres in size (FIGURE 3.39 E , F ) . Olivine macrocrysts are included in juvenile pyroclasts or are present as discrete grains. These grains are typically sub-rounded to ovoid, highly fractured and altered to serpentine and calcite. They range in size from 2-5mm. Olivine macrocrysts occur in all lithofacies except L3 where they are only noted on one surface (FIGURE 3.38D). In L2 they typically occur as angular fragments rather than whole rounded grains (FIGURE 3.37 B , C , & D ) . Olivine phenocrysts are abundant both within juvenile pyroclasts, where they make up 20-25%, and as loose clasts. 99 These grains are typically euhedral, elongate prisms (FIGURE 3.36D). In L2, they occur as angular fragments rather than euhedral crystals. In L3, they occur in distinct beds, but they are not present in all beds (FIGURE 3.38E). They phenocrysts are much smaller in L3, where they are <lmm in size. Country rock xenoliths have irregular distribution. They are typically angular to sub-angular mudstone xenoliths. More irregular mudstone clasts (FIGURE 3.39B) are less common, as are sub-rounded carbonate xenoliths and rare granitic to dioritic basement clasts. Mantle xenoliths include rare ~1 cm peridotite xenoliths, consisting mainly of olivine with occasional garnet and chrome diopside. Colloform or crystalline carbonate is the dominant interclast fill in all lithofacies except L3 where cryptocrystalline carbonate is the dominant material. Colloform carbonate is the dominant fill especially in coarser beds (FIGURES 3.36D and F; FIGURES 3.39 B,C, and D) and commonly is incomplete fill as there are open pore spaces between the grains. In L3 matrix consists of finely laminated carbonate (FIGURE 3.38C,D,E&F). Structures and Sorting Fine laminar, horizontal to subhorizontal bedding is the dominant structure in lithofacies 1 and parts of lithofacies 2. Bedding thickness ranges from mm to dm scale (FIGURE 3.36A-C). The highest angle beds are around 57'tea. Coarser beds have average grain sizes up to 7 mm and dominantly contain juvenile pyroclasts. Finer beds have average grain sizes between 0.5 and 1 mm and are composed mainly of olivine crystals; Cross-stratification is present sporadically throughout LI but is subtle due to the thinness of bedding (FIGURE 3.36A). Rare, normally graded sequences are present (FIGURE 3.36B). Massive intervals are common in L2 (FIGURE 3.37A) and dominate L4 (FIGURE 3.39A). L3 is dominated by fine laminations (FIGURE 3.38A,B). . 100 Overall lithofacies 1,2 and 4 are enriched in olivine crystals relative to estimates of the parent magma based on the juvenile pyroclasts (FIGURE 3.36E). This suggests that these deposits were sorted and that fine matrix material has been lost. Individual beds are relatively well sorted however erratic large fragments do also occur. Summary and Interpretations L I consists of well-sorted juvenile pyroclasts and olivine crystals that show little to no evidence of abrasion. As a result these deposits likely represent primary pyroclastic deposits. There are significant grainsize variations between the layers of the same cross-bed set, which is typical of pyroclastic surges due to their pulsating, non-equilibrium character when compared with epiclastic processes that generate cross-beds (Cas and Wright, 1987). In contrast, L 2 contains abundant broken clasts and is depleted in juvenile pyroclasts relative to L I and L 4 . This suggests epiclastic reworking of the pyroclastic kimberlite where juvenile pyroclasts were abraded and destroyed and olivine grains broken and concentrated. L 2 is often massive with rare bedding, and is intensely altered. These characteristics are consistent with epiclastic reworking. L 3 is composed of fine cryptocrystalline carbonate that may represent altered kimberlite ash. The fine carbonate laminae could represent a hiatus in eruptive activity where fine carbonate rich kimberlitic ash was able to settle out along with fine phenocrysts. The single bedding plane that is defined by juvenile pyroclasts and olivine macrocrysts represents a distinct event, most likely a small eruption that contributed no new material, within a period of low energy sedimentation. L 4 is similar to L I in composition however it lacks the structure of L I . As a result L 4 is interpreted as proximal pyroclastic fall and synvolcanic slump deposits, occurring as intra-crater fill. The presence of accretionary pyroclasts, although rare, suggests that some of the material 101 has been transported in a dilute column or cloud and this suggests that LA was deposited as pyroclastic fall or other proximal pyroclastic deposits. Overall the deposits and the suggested geometry from drillcore intersections, suggests that body K296 is composed of deposits that transition from vent proximal pyroclastic fall deposits to crater edge mixed surge and fall deposits. Overlying these pyroclastic deposits are reworked pyroclastic kimberlite (FIGURE 3.35). 102 3.4.6 K300 The three drillholes into body K300 were all available. Two of the drillholes contained substantial intervals of kimberlite however the third intersected only 1.3 m. The surface expression as inferred by geophysics is presented in figure (FIGURE 3.40). The main features of body K300 are summarized in T A B L E 3.9. Geometry The deposits of body K300 thin rapidly from a thickness 60.6 m in the north to a thickness of 1.3 m in the south over a 200 m interval (FIGURES 3.40 - 3.43). This material could be interpreted as either a shallow crater edge or a ring of volcanic material (FIGURE 3.44). Lithofacies Four lithofacies are identified in body K300. Lithofacies 1 (LI) consists of mixed kimberlite and mudstone and is the only kimberlitic material in hole K300-02. LI also forms the upper part of the kimberlite in hole K300-03 (FIGURE 3.42c, 3.43c). Lithofacies 2 (L2 ) is a fine-bedded, olivine crystal-rich, juvenile pyroclast-bearing kimberlite (FIGURE 3.45A-B). This lithofacies comprises a sequence of cm scale beds that alternate coarse and fine with no gradation between individual beds. Bedding angles are generally flat lying; however locally there is considerable variation over scales of tens of centimeters. Cross-bedding is common. Lithofacies three (L3) consists of massive, olivine crystal-rich, juvenile pyroclast-bearing kimberlite (FIGURE 3.45C-D). Lithofacies four (L4) consists of a thin interval of densely packed olivine crystal-rich kimberlite (FIGURE 3.45E,F). Lithofacies 1,2 and 3 are found in gradational contact with one another whereas L4 occurs as an isolated intersection of kimberlite encased in mudstone. Interclast matrix in all units is composed predominantly of carbonate with areas of 103 Figure 3.40: A. Gravity anomaly expression of body K300 showing distribution of drillholes and inferred (bold lines) surface expression from electromagnetic data. B. East-facing cross sectional view of drill holes showing distribution of kimberlite, host sedimentary rocks and overlying Quaternary sediment. All holes shown were examined in this study. 104 Table 3.9: Characteristics of kimberlite deposits in body K300 Lithofacies Geometry Clast Population Matrix/ Interaranular fill Structures Alteration Interpretation Deposit type LI: Mixed kimberlite and mudstone Occurs as a thin (<lm) horizon within an interval of mudstone JP AP -OM OP CX MX -- P N/O - P - P - P N/O Mudstone showing soft sediment features Massive Mixed material Intensely characteristic of Reworked altered reworked pyroclastic material kimberlite L2: Fine-bedded, olivine crystal-rich, juvenile pyroclast-bearing kimberlite Laterally continuous. Thins from 36 to Om over a distance of 200m JP - P AP - N/O OM - A O P - A CX - P MX - N/O Microcrystalline carbonate. Mix of cryptocrystalline carbonate, serpentine and other clay minerals Finely bedded Intensely altered. Olivines altered to carbonate and serpentine Fluctuating energy levels Pyroclastic Surge L3: Massive, JP AP -- P Microcrystalline Massive olivine crystal- Laterally N/O carbonate. Mix of overall. Subtle Subtle gradations rich, juvenile continuous in OM - A cryptocrystalline grainsize Intensely without distinct beds Pyroclastic pyroclast- two holes but OP - A carbonate, gradations but altered indicates a fluctuating fall bearing tapers out CX - p serpentine and other no distinct environment. kimberlite N/O clay minerals beds MX -L4: Densely Laterally continuous but kimberlite holes 300-01 and 300-03 JP - N/O AP - N/O OM - A OP - A CX - P MX - N/O Microcrystalline carbonate. Mix of cryptocrystalline carbonate, serpentine and other clay minerals Generally massive with grain orientation parallel to contacts Intensely altered Dense packing, concentration of heavy minerals Reworked pyroclastic kimberlite Clast Types: JP= juvenile pyroclast, AP= accretionary pyroclast, OM = olivine macrocryst, OP = olivine phenocryst, CX = country rock xenolith, MX = mantle xenolith Clast abundance: A=abundant, P=present, R=rare, N/0= not observed 105 MEAN GRAIN SIZE ash lapilli F MC F M C UIIII Maximum Clast Size (mm) Percentage Clast Type 0 2 5 10 20 50 > JP AP LC OM/OP 0 Rock Type IR Lithofacies Interp D ~ — . . 0 • • —, 0 • • • . — ° o'<i'. o ' -D — 0:-::-;.°;:-; .9/ —. . 0 — ° •'.-7-'. '•>•i'. °"rr' • "o. a' . 3 — ". ~ P ' •.• • . ° . ' . 0 : . p . q . 10 55 35 Clay rich till Sandy Mudstone Contains fossil shell material Fine-bedded kimberlite Massive kimberlite Mudstone L2 L3 PS PF/PS 100 125 150 0. 60 Olivine crystal-rich kimberlite L4 RVK Figure 3.41: Stratigraphic section for vertical hole K300-01. Grain size: F= fine, M= medium, C=Coarse. Clast type: JP= Juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst, 0=opaques, P=phlogopite. IR = Interval reference. Interp= Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode of deposition: S= normal sedimentation, PF=pyroclastic fall, PS= pyroclastic surge. RVK = reworked volcaniclastic kimberlite. EOH= end of hole. 106 MEAN GRAIN SIZE ash lapilli F M C F M C I I 1 1 1 1 Maximum Clast Size (mm) Percentage Clast Type 0 2 5 10 20 50 > JP AP LC OM/OP 0 Rock Type IR Lithofacies Interp . ° c . " - ° - 7 ^ - ; -•'.°°'-r-.'-o-'<i'' • o • • . O ' Clay rich till Sandy Mudstone Sandy Mudstone zzozz; Figure 3.42: Stratigraphic section for vertical hole K300-02. Grain size: F= fine, M= medium, OCoarse. Clast type: JP= Juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst, 0=opaques, P=phlogopite. IR= Interval reference. /nferp= Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode of deposition: S= normal sedimentation, RVK = reworked volcaniclastic kimberlite. EOH= end of hole. 107 MEAN GRAIN SIZE ash lapilli F M C F M C | | | | | | 0 Maximum Clast Size (mm) Percentage Clast Type 2 5 10 20 50 > JP AP LC OM/OP 0 Rock Type IR Lithofacies Interp 25 50 75 100 125 150 .V." .o -T^ . l : . • .eJ- 'O- ' . - lo. ' . - . ' . -.o'.-. '-d-10 60 60 <2 2-5 Clay rich till Sandy Mudstone Kimberlite/ Mudstone Fine-bedded kimberlite Fine Bed Coarse Bed Massive kimberlite There is some interlayering of kimberlite and mudstone at contact Mudstone Y Olivine crystal-rich kimberlite Mudstone L I L2 L3 L4 RVK PS PF/PS RVK Figure 3.43: Stratigraphic section for vertical hole K300-03. Grain size: F= fine, M= medium, C=Coarse. Clast type: JP= Juvenile pyroclast, AP=accretionary pyroclast, LC= lithic clast, OM/OP = olivine macrocryst/olivine phenocryst, 0=opaques, P=phlogopite. IR= Interval reference. /nferp= Volcanological interpretation of dominant mechanism of deposition for each lithofacies. Mode of deposition: S= normal sedimentation, PF=pyroclastic fall, PS= pyroclastic surge. RVK = reworked volcaniclastic kimberlite. EOH= end of hole. 108 300-03 125 175 300-01 . . . . 2 5 " e.*:«*v«. '•••MA it V L4 Reworked pyroclastic kimberlite 150 175 200 Till Mudstone 300-02 o- • 1 ^ 50 -1* Mudstone 125 - t = . 7 d 175 Figure 3.44: Proposed geometry and deposits of body K300 based on logging of holes 300-01,300-02 and 300-03. 109 Figure 3.45: A. L2 consisting of finely bedded deposits. Note the scour mark at the base of the light grey bed. B. L2 in hole 300-03 with distinct bedding defined by clast size and clast orientation. C. Contact between underlying massive L3 and overlying bedded L2 (black dotted line). D. L3 which is massive to very subtly bedded. E&F. L4 in holes 300-01 and 300-03 respectively. Olivine crystal-rich with bedding defined by clast orientation. White arrows indicate stratigraphic upward facing direction. 110 cryptocrystalline carbonate and serpentine. Areas within the body are pervasively altered light grey to brown, vuggy and smell strongly of bitumen. Patchy alteration gives the rock a mottled appearance and obscures primary textures and structures. Late veins of quartz and bitumen occur randomly throughout the body. Alteration of body K300 ranges from texturally destructive to pervasive but texturally preservative. Clast Population The clast population for K300 includes juvenile pyroclasts, olivine macrocrysts, olivine phenocrysts, country rock xenoliths, mantle xenoliths and interclast matrix/fill. Juvenile pyroclasts in L2 and 3 constitute up to 10% of the material present. Juvenile pyroclasts are typically 0.5-10 mm in size and occur as both nucleated and non-nucleated bodies. They are typically amoeboid in shape (FIGURE 3.46A,B), contain variable amounts of olivine and phlogopite, and have a carbonate-rich groundmass. The rounded to amoeboid juvenile pyroclasts show no evidence of abrasion. Pyroclasts are typically non-vesicular however rare vesicles are observed (FIGURE 3.46C,D). Juvenile pyroclasts are not observed in L4 and are rare in LI. Accretionary pyroclasts are not observed. Olivine macrocrysts and phenocrysts are the dominant clasts in L2 and L3 and constitute up to 70% of the rock (FIGURE 3.46E,F). Macrocrysts are typically 1-5 mm, rounded, fractured, and altered to carbonate, although rare fresh kernels are present. Olivine phenocrysts are typically <0.5 mm, and tend to be more euhedral than macrocrysts. In LI, olivine crystals constitute around 20% of the material present. In L4, >2mm, olivine crystals comprise more than 60% of the rock. Phlogopite occurs in all units but with variable distribution. Locally it can make up to 15% of the material present. It typically occurs as small books <0.5 mm across. Country rock xenoliths make up 2-10% of the body and are mainly black subangular, shale fragments ranging in size from 2 mm-3 cm. White to light pink, subangular, carbonate 111 Figure 3.46: A. Juvenile pyroclast nucleated on olivine macrocryst and containing olivine phenocrysts and phlogopite. B. Non-nucleated, juvenile pyroclast with fluidal shape containing olivine macrocrysts phenocrysts and phlogopite lathes. C. Broken juvenile pyroclast containing a single carbonate lined vesicle. D. Enlargement of juvenile pyroclast shown in C. Photomicrograph is in crossed polars. Note carbonate lined vesicle and phlogopite lathes. E. Large olivine macrocrysts. F. Very well sorted, olivine phenocryst-rich deposit. Note the euhedral grain shapes shown in the inset in the upper right hand corner. Abbreviations:OM = olivine macrocryst, JP = juvenile pyroclast, Phi = phlogopite,V = vesicle, OP = olivine phenocryst. 112 xenoliths tend to be slightly larger ranging from 5 mm-4 cm. Rare granitic xenoliths locally have thick dark alteration rims. Rare quartz grains are intermixed with kimberlite. In LI about 50% of the rock is mudstone, which is intermixed with kimberlite. Soft sediment deformation textures indicate that the mudstone was soft at the time of deposition. Chromite, chrome rich diopside and garnet with kelyphite rims are present, but in low abundance. Structures and Sorting L2 beds are typically subhorizontal, 80-90° to core axis, but can be as steep as 70° to core axis. Bed thickness is centimetre scale (FIGURE 3.45A,B) and defined by grain size variation and by orientation of elongate grains. Individual coarse and fine-grained beds alternate on a cm-scale and are well sorted within the beds. Some of the thicker beds are normally graded. Cross-bedding is common although not pervasive. LI is discontinuously bedded with intermixed kimberlite and mudstone. Bedding is poorly defined. Bedding is not observed in L3 and 4 (FIGURE 3.45C,D,F). Summary The thick accumulation of kimberlitic material observed in holes K300-01 and K300-03, thins rapidly, resulting in a profile that is inconsistent with normal sedimentation as the angle of deposits is greater than the expected angle of repose. This indicates a pyroclastic origin for these deposits. The clast population also supports an interpretation of pyroclastic as most grains show little or no evidence of abrasion. The abundance of olivine crystals indicates crystal concentration in these deposits that could occur through the elutriation of fines in a volcanic cloud. From the intersections examined, body K300 could represent a positive relief feature such as a tephra ring or a negative relief feature such as a crater. Of the three holes through K300, two show gradational contacts between overlying mudstone and kimberlite and one is in sharp 113 contact with overlying sedimentary rocks. Lower contacts with mudstone are typically gradational however in K300-03 the contact is sharp, but interlayering between kimberlite and mud is observed. From the above considerations it would seem that the kimberlite was probably deposited much more rapidly than the encasing shale and sandstone. However in the initial phases a higher proportion of country rock was included in the deposits. The top of the deposits is likely reworked. The initial phases are gradational and may result from incorporation of underlying unconsolidated shale. Upper gradational contacts represent likely represent epiclastic re-working of kimberlite. L2 is interpreted as pyroclastic surge deposits as indicated by the geometry of the deposits and non-abraded grains. Another consideration is that the material present is almost exclusively of kimberlitic origin with little country rock contamination except for at the base and tops of the unit. The repetitive alternation of coarse and fine beds is commonly observed in surge deposits and reflects the pulsating nature of surge deposition. Very fine ash laminae could result from co-surge ash fall as some material is elutriated from the surge and falls out later. Surges commonly include features such as cross bedding, and can form thick accumulations that are not laterally extensive. LI consists of intermixed kimberlite and mudstone and is interpreted as a reworked or distal equivalent of L2. L3 is relatively unsorted and massive, but otherwise similar to L2 and as a result is interpreted as deposition of material from mass flow. Its features are consistent with debris flows, pyroclastic flows or slumping. The main characteristics are that it lacks any distinctive sorting or structures and therefore could be representative of any process that deposits volcanic material without sorting. L4 is a concentrated crystal-rich deposit that has higher quantities of garnet and olivine than the other lithofacies examined. No juvenile or accretionary pyroclasts are observed. As a result L4 is interpreted as resedimented pyroclastic kimberlite where heavy minerals have been concentrated. This unit is inferred to be associated with an older event, as it occurs at depth 114 within the hole and is separated from the other kimberlite intervals by a substantial (>40 accumulation of mudstone. 115 3.5 Summary A summary of the main features of bodies K6 , K l l , K252, K281, K296, and K300 is presented in Table 3.10. In all deposits, crystal contents are high compared with analogous deposits from other magma types. Accretionary pyroclasts are very abundant in bodies K252, and K281, occur in bodies K296 and K6 , and are not observed in bodies K300 and K l l . Juvenile pyroclasts are abundant in bodies K296, K l 1 and K281, and are present in bodies K300, K 6 and K252. Bimodal clast populations are present in several bodies. In body K252 accretionary pyroclasts and country rock xenoliths occur in a coarse size fraction while olivine crystals and juvenile pyroclasts occur in a medium size fraction. Similarly in body K281 accretionary pyroclasts and country rock xenoliths occur as course clasts whereas olivine crystals and juvenile pyroclasts are medium to fine-grained. In body K296, coarse beds are dominated by juvenile pyroclasts whereas fine beds consist dominantly of olivine crystals. Interclast matrix ranges from colloform textured carbonate, to microcrystalline mixed carbonate and serpentine to translucent isotropic serpentine. The colloform textured carbonate clearly represents open space fill however the origins of the latter two are enigmatic. Possibilities include replacement of fine ash, hydrothermal fluids and alteration, and precipitation from magmatic fluids (Webb et al., 2003). The features and interpretations recorded indicate that each of the bodies is representative of a slightly different zone within a kimberlite. Body K 6 is interpreted as proximal crater fill from pyroclastic fall, with rare intermittent surge events recorded by discrete intervals. Body K l 1 is also interpreted as proximal crater fill however no surge events are recorded in this interval. Body K281 is modeled as interstratified pyroclastic fall and surge deposits where surge events compose a' substantial proportion of the deposits. Body K296 contains horizons of more massive pyroclastic fall and slumping as well as more distal intersections dominated by surge 116 Table 3.10: Summary of features of bodies 6, 11, 252, 281, 296, and 300 Body Geometry Deposits Defining features K6 K l l K281 K296 K300 K252 Vertically continuous intersections. Intra crater deposits. Vertically continuous intersections. Intra crater deposits. Vertically continuous intersections. Intra crater deposits. Significant extra crater pile transitional to intra-crater, vertically continuous sequence (to depth of drilling). Thins from thick sequence (>50m) to absent. Interpreted as extra-crater or shallow crater edge deposits Multiple layers within a sequence of mudstone. Extra-crater deposits Proximal pyroclastic fall and slumping, possible intermittent surge Proximal pyroclastic fall and slumping Pyroclastic fall and pyroclastic surge Pyroclastic fall and pyroclastic surge. Structures become better developed with distance from vent Resedimented on top Pyroclastic flow, pyroclastic surge, and resedimented kimberlite Pyroclastic fall and resedimented kimberlite Crystal-rich deposits Fresh olivine Massive Carbonate and mixed carbonate/serpentine matrix Crystal-rich deposits Fresh olivine Massive Serpentine and mixed serpentine/carbonate matrix Recycled pyroclasts Crystal and accretionary pyroclast-rich Mixed serpentine carbonate matrix Recycled clasts Transition from massive to well bedded. Colliform carbonate matrix Crystal and Juvenile pyroclast rich Massive basal unit Overlain by finely bedded crystal rich deposits. Crystal and juvenile pyroclast-rich Large accretionary pyroclasts and bimodal clast population in pyroclastic fall. Well-sorted and abraded olivine crystals in resedimented kimberlite. 117 deposits. Body K300 is similar and consists of a basal massive unit that could be a pyroclastic flow or fall followed by overlying surge deposits which are considered transitional from intra to extra-crater deposits. Finally, body K252 has many temporally separated intervals of pyroclastic fall and reworked pyroclastic kimberlite. This material is interpreted as extra-crater deposits, lying outside of the main volcanic edifice. In the next section the commonalities and differences elucidated here will be used to construct a geometric and theoretical model for the appearance and eruptive mechanisms of an ideal Buffalo Head Hills kimberlite. 118 Chapter 4: Composite volcanological model 4.1 Rationale for the composite model In Chapter 1,1 presented volcanic facies from six kimberlite sequences and proposed models for the deposit types and geometries of kimberlite in the BHH region (FIGURE 4.1 A-F). Each model is based on the specific characteristics observed in each location, including geometry, clast population, sorting and structures. No single body contains all features observed in this region. However comparison of stratigraphic sequences of individual volcanoes shows a common set of deposit types. None of the bodies have been drilled sufficiently to give a picture of an entire kimberlite volcano but an 'average' kimberlite volcano from the Buffalo head Hills can be conceptualized by combining the stratigraphic relationships into a composite model. The first consideration when constructing a composite model stratigraphy, is whether all of the observed sequences derive from a single style of volcanic eruption or whether they require multiple styles (e.g. Hawaiian vs. Plinian). At BHH all kimberlite bodies have characteristics suggesting deposition within a volcanic crater and/or associated tuff ring. Pyroclastic fall deposits are ubiquitous, and pyroclastic surge deposits are common. The clast populations for all bodies are similar, although the proportions vary within and between bodies. On this basis, I consider that the BHH kimberlite deposits have been deposited by broadly similar eruptive mechanisms. 4.2 Composite volcanological model The facies and deposit distributions from six kimberlite bodies are compiled to generate an image of a single kimberlite volcano (FIGURE 4. IG). The idealized kimberlite volcano, shown 119 Figure 4.1: Proposed geometr ies for A . body K6, B. body K l 1, C. body K296, D. body K281, E. body K252, F. body K300 and C a n ideal ized k imber l i te body in the B H H . Abbreviations: PF=pyroclast ic fall, PS=pyroclast ic surge, RVK= reworked volcanic last ic k imber l i te . 120 in FIGURES 4.1G and 4.2, comprises a large crater partially infilled by pyroclastic material and surrounded by a pyroclastic tephra ring. More distal pyroclastic fall deposits are found further out from the vent. Late epiclastic material also occurs as crater fill, and overlies the pyroclastic fill (Figure 4.1g, 4.2). This model represents an idealized kimberlite volcano in the BHH area prior to significant erosion or burial. Below, I review the main elements of this model stratigraphy. Geophysical data suggests that the average surface expression for a BHH kimberlite volcano is 500 m in diameter. Crystal-rich vertically continuous primary pyroclastic fall (bodies K6 and K l 1) fills the vent of the kimberlite (FIGURE 4.1A&B, 4.2A). Towards the edges of the crater are facies where the proportions of juvenile and accretionary pyroclasts increase and surge deposits become common (bodies K281and K296, FIGURE4.1C&D, 4.2B). Extra-crater deposits are dominated by surge and pyroclastic flow deposits (FIGURE 4 . I F & 4.2c, body K300) and distal deposits include pyroclastic fall and epiclastically reworked kimberlite (body K252, FIGURE 4. 1E & 4.2D). Original differences in relief between tephra ring and crater are minimized by epiclastic reworking which acts to erode the tephra ring and infill the crater (FIGURE 4.2E). Reworked deposits of this nature are present in bodies K296, K300 and K252. Pyroclastic kimberlite becomes progressively fresher with proximity to the vent as a result of the more rapid cementation of vent proximal deposits due to heat and abundance of fluids. Translucent serpentine matrix is found exclusively in these deposits and its presence may be heat dependant. Vent proximal deposits are enriched in olivine crystals and depleted in juvenile pyroclasts in comparison with crater and extra crater deposits. This could result from density sorting because denser olivine crystals were deposited close to the vent, and lower density juvenile pyroclasts being concentrated'farther out. Deposits become better sorted with distance from the eruptive centre, and surge deposits are more common towards the edge of the crater and in crater rims. The re-worked pyroclastic kimberlite consists mainly of abraded and 121 TVS . A. Vent proximal crystal-rich pyroclastic fall (K11 and K6) B. Vent proximal mixed crystal-rich and accretionary pyroclast-rich pyroclastic fall with minor surge (K6 and K81) C. Mixed pyroclastic fall and surge deposits, intra-crater but not vent proximal (K296) D. Pyroclastic Surge dominated intra and extra-crater deposits (K300) E. Extra-crater pyroclastic fall and re-worked pyroclastic kimberlite (K300, K252) F. Extra-crater deposits from previous eruptions (K252) Figure 4.2: Schematic three dimensional model of a Buffalo Head Hills kimberlite 122 broken olivine crystals. These rocks are depleted with respect to juvenile pyroclasts and accretionary pyroclasts. Where these clasts are present they are commonly abraded. The epiclastic deposits occur distally, in the volcaniclastic apron and fill the residual crater depressions. 4.3 Relevance of the volcanogenic model to eruptive style The features of BHH kimberlites are considered in order to distinguish the possible eruptive mechanism for kimberlite in this region. Firstly, the study area is underlain by 325 m of Cretaceous sediments. All evidence suggests these sediments were water saturated and unconsolidated at the time of eruption. This situation strongly supports an important phreatomagmatic component to kimberlite eruption. Furthermore, the kimberlite may have erupted into a standing volume of water. The main deposit types identified in the BHH kimberlites are pyroclastic surge, pyroclastic fall, massive proximal pyroclastic deposits, and reworked pyroclastic deposits. The reworked pyroclastic and massive pyroclastic deposits provide little direct information about volcanic processes. Pyroclastic fall deposits indicate the presence of a volcanic column supported by continuous to semi-continuous uprush of a dispersed flow of pyroclasts gases and liquids. Accumulations of pyroclastic fall are thickest close to the vent and become thinner and better sorted with distance from the vent. Near-vent pyroclastic fall deposits are enriched in crystals and depleted in fines, which suggests elutriation of fines in the volcanic column and preferential sedimentation of crystals due to their high density. Pyroclastic fall deposits have the widest distribution of the pyroclastic deposits identified and occur as discrete horizons outside of the tephra ring. 123 Pyroclastic surge deposits occur mainly in the tephra ring, and to a lesser extent as crater fill. Surges are defined as low concentration pyroclastic density currents and suggest an abundance of gas and condensing liquid phases which can suspend significant pyroclastic material. The presence of these deposits is a strong indication that the ascending kimberlite magma interacted phreatomagmatically with surrounding wet sediments or water. Pyroclastic surge deposits are typically associated with phreatomagmatism; however, they are not restricted to this style of eruption. For example, Valentine and Fisher (2000) noted surge deposits derived from eruptions that were driven solely by volatile exsolution. There are three main particle types that provide evidence of process in these deposits. These are accretionary pyroclasts, juvenile pyroclasts and discrete olivine crystals. Accretionary pyroclasts indicate the presence of water which allows individual ash particles to accrete and form large spherical bodies (Sparks et al. 1997, Gilbert and Lane, 1994). Accretionary pyroclasts are very commonly associated with phreatomagmatism (Cas and Wright 1987) and are used as an indicator that water was involved in the eruption. This supports the assertion that kimberlite eruption in the BHH was, at least in part, phreatomagmatic. Juvenile pyroclasts present in Buffalo Head Hills kimberlites are typically fluidal-shaped, non-vesicular bodies. These clasts are not consistent with typical juvenile pyroclasts generated from exsolution driven eruptions from basaltic magmas. In basaltic systems, juvenile pyroclasts are typically highly vesicular where fragmentation results from gas exsolution. The absence of vesicles is typical of kimberlitic juvenile pyroclasts in most observed localities (Scott-Smith, pers.comm.). Clement (1982) attributes the absence of vesicles in kimberlite juvenile pyroclasts to either extensive degassing of the magma prior to explosively induced disruption or to the attributes of the low magma viscosity, which facilitates effective degassing (Clement, 1982; Dawson, 1980). Other explanations for the lack of vesicles in juvenile pyroclasts include the effectiveness of C 0 2 as a gas sparging agent (Rice, 1999), the difficulty in preserving vesicles in 124 fragile magmas that do not easily form glass, and readsorption of volatiles into crystallizing carbonate phases. A further consideration is the intense alteration of juvenile pyroclasts in the BHH region that may erase any fine vesiculation as well as possible larger vesicles. Another explanation for the lack of vesicles in juvenile pyroclasts is that the kimberlite magmas were not vesiculating at the time of eruption. Non-vesicular fluidal to spherical pyroclasts have been generated experimentally by Lorenz et al. (1991) through phreatomagmatic interactions. However, Zimanowski (1998) states that although non-vesicular, fluidal-shaped juvenile pyroclasts can be produced by these interactions, they are not diagnostic features of phreatomagmatism. The fluidal shapes do suggest a low viscosity melt phase that was able to relax to a fluidal shape after fragmentation. A further implication of the preserved shapes is that the juvenile pyroclasts had to be relatively cool before deposition. The pyroclasts do not record deformation as a result of impact or after deposition, with rare exceptions (FIGURE 3.6A). The primary mineralogy of Buffalo Head Hills kimberlites has not been investigated in this study, however some general comments can be made regarding the presence of abundant serpentine and carbonate. Most of the deposits examined in the BHH kimberlites are extensively altered to serpentine and carbonate mineral-rich assemblages. Extensive examination of the alteration assemblages has not been undertaken however it is likely that much of this alteration was synvolcanic and resulted from kimberlite-derived fluids. As a result it is reasonable to assume that the ascending kimberlite magma contained a significant amount of dissolved volatiles, and that exsolution was occurring to some degree. Summary The eruptive products and physical volcanology of BHH kimberlites are, in many aspects, consistent with both phreatomagmatic and exsolution driven eruption. Evidence for 125 phreatomagmatic eruption of kimberlite is ubiquitous in BHH kimberlites. Firstly, the paleo-environment during eruption is ideal for phreatomagmatism. Secondly, many aspects of the pyroclastic kimberlite deposits are suggestive of magma-water interaction. Therefore it is clear that kimberlites in the BHH erupted phreatomagmatically; however, this does not mean that magma-water interactions were the driving force for kimberlite eruption. Kimberlite, and every other magma, is driven to surface as a result of positive buoyancy. This positive buoyancy can be enhanced by the exsolution and expansion of volatiles species. Most studied kimberlites have been modeled as having high volatile content, and that volatile exsolution plays a large role in kimberlite eruption. Volcanological studies of BHH kimberlites have not provided direct evidence to support exsolution of volatiles; however, it is considered likely that exsolution of volatiles played a significant role in kimberlite volcanism. Carbonate is the dominant alteration mineral in BHH kimberlites, and this reflects a significant volatile component to the kimberlite magma. In the next chapter, previously proposed mechanisms for kimberlite ascent and eruption are reviewed, and a new model is proposed for the eruption of kimberlite in the BHH area. 126 Chapter 5: Eruptive mechanisms Kimberlite eruption is the result of magma ascent from mantle depths in excess of 200 km that is sufficiently fast to preserve diamond. There have been no modern kimberlite eruptions and, thus there is debate concerning the styles and mechanisms of kimberlite eruption. Two main models have been proposed for eruption of kimberlite. The first model, the exsolution model, asserts that kimberlite characteristics result from the intrusion of volatile rich magmas which evolve into a fully fluidized gas-solid-melt system (e.g. Clement 1982, Clement and Skinner 1979, Clement and Reid 1989; FIGURE 5.1 and 5.2). An alternative model, the phreatomagmatic model, asserts that explosive kimberlite volcanism results from interactions between kimberlite magma and groundwater (Kurszlaukis et al. 1998). This results in progressive excavation, and filling of the pipe, by downward subsidence of material into the crater/diatreme (e.g. Lorenz et al. 1991, Lorenz 1987, Lorenz and Kurszlaukis, 2003; FIGURES 5.3,5.4 and 5.5). Neither of these two end-member models suffices to explain the deposits found in BHH kimberlites. Here I propose an integrated volcanological model for the eruption of BHH kimberlites. This model has the following basic elements: a.) Initiation of ascent of a kimberlite magma with some reasonable initial volatile content; b.) decreasing pressure associated with magma ascent results in saturation of the magma, and nucleation of immiscible fluid droplets; c.) droplet growth which results in expansion, upward acceleration of flow and eventual fragmentation of magma to form a dispersed flow; d.) intrusion into wet sediments and onset of phreatomagmatic interaction; e.) surface breach, eruption, and downward propagating decrease in pressure; and f.) continuous to semi-continuous flow of material from the vent, formation of a volcanic plume and initiation of pyroclastic surge events. The main attributes of this model are that it explains the observed volcanic facies, it allows for the presence of reasonable initial volatile contents that result in vesiculation and 127 EPICLASTICS Figure 5.1: Model of an idealized kimberlite magmatic system illustrating the relationships between crater,diatreme,and hypabyssal facies rocks (not to scale). Hypabyssal facies rocks include sills,dikes, root zone and "blow".Modified from Mitchell (1986) after Dawson (1971), Hawthorne (1975), Clement and Skinner (1979), and Scott Smith and Skinner (1984). 128 Figure 5.2: Schematic representation of emplacement model for southern African kimberlites. From Field and Scott Smith (1998) after drawings by C.R.Clement (1982), Clement and Reid (1989), and Field et al.(1997).V»o6reW0f/ons:TKB = tuffisitic kimberlite breccia, HK = hypabyssal kimberlite.a,b,c Kimberlite magma intrudes along jointing in the country rock and rises slowly by stoping and wedging volatiles migrate to the head of the magma gas column. Where the rising magma meets a barrier, there is a temporary halt in ascent and build-up of the gas cap. Eventually, pressure build up and fracturing result in breaching of the barrier by explosive brecciation and upward migration of the magma continues. The black arrows in c represent the forcing of volatiles into country rock. d. Explosive breakthrough occurs when the volatile pressure exceeds the confining pressure of the last barrier. A crater is excavated and the magma column begins to degas.The volatiles forced into the country rock now migrate back inward resulting in wallrock brecciation. e. Carbon dioxide and other volatiles continue to exsolve resulting in fluidisation of the magma. f.The system cools rapidly. 129 Figure 5.3: A 3D block diagram of possible kimberlite volcanic landforms. Note that the root zones may have different depths, depending on the magma/water reflux rate. Modified from Kurszlaukis and Lorenz (2003). 130 Maar with lake Figure 5.4: Schematic representation of the evolution of a maar-diatreme volcano assuming restricted availability of groundwater and the formation of a cone of depression (stippled line). Because hydrostatic explosions are pressure dependant the diatreme penetrates downward with time, leading to a diatreme 2000- 2500 m deep. X is the assumed maximum depth of groundwater column on water vapour explosion site. After eruptions cease, the groundwater table may restore itself to original levels, usually leading to formation of a maar lake. Modified from Lorenz (1986). 131 Figure 5.5: A.The root zone and lower diatreme directly after an explosion. The country rock is intensely brecciated and the explosion chamber is evacuated by the eruption through the feeder vent. A short lived cavity forms at the explosion site. B. The cavity is filled by rockfalls and rockslides from the unstable brecciated side walls and also by material subsiding from higher explosion chamber(s) and the overlying diatreme. Mag ma (red) intrudes the contact breccia and water enters the vent. C. Situation immediately after a new explosion. Highly energetic shock waves are emitted by the explosion and brecciate the country roclcThe volume expansion due to steam generation pushes the material overlying the explosion centre upwards. D.The expanding vapor creates a magma-solid-water-vapor system that expands against the over-lying diatreme-filling tephra, through which it will pierce to form a feeder vent ejecting material to the atmosphere.The explosion chamber will be evacuated to create the situation pictured in A. During the cyclic phases of explosion, eruption, collapse and intrusion, the root zone and the diatreme penetrates downward and grows radially outward. E. One of the possible end conditions after volcanic activity ceased. Magma ascent stops and a hydrothermal system is set up as it cools. F. Second possible end condition where the system 'ran dry'.The water reflux stopped while magma ascent continues. Modified from Kurszlaukis and Lorenz (2003). 132 disruption of kimberlite magma, and it includes the role of groundwater in the amplification of kimberlite eruption processes. Prior to presenting the eruptive sequence envisioned for BHH kimberlites, previous models are reviewed and assessed. 5.1 The Exsolution Model The exsolution model is based predominantly on textures observed in South African kimberlites (Clement 1982, Clement and Skinner 1979, Field et al. 1997), although examples of similar kimberlite textures have been noted in Canada (Herman, 2003). In South Africa, erosion of the kimberlites is significant, with as much as 900 to 1900m removed (Clement, 1982). The consequence is that the model is based predominantly on diatreme and hypabyssal rocks because the extrusive facies are not present. This model does however postulate the presence of large maar-like craters and associated tuff rings forming the top of the system (FIGURE 5.1) as blocks of pyroclastic and epiclastic deposits occur as downrafted blocks within diatremes. Furthermore, in other regions of Africa, kimberlite craters are observed (Tanzania, Zambia, Botswana, Angola, Zaire, Mali; Tremblay, 1956, Mannard 1962, Edwards and Howkins, 1966, Hawthorne, 1975). The exsolution model proposes a kimberlite eruption style that begins with embryonic kimberlite pipes migrating upwards through crustal rocks, and culminates with explosive volcanic eruption that is driven by exsolution of volatiles (Clement, 1982). Clement (1982) also allows for the possibility that explosive eruption can, in part, be phreatomagmatic. Before breaching the surface, hydraulic fracturing, magmatic stoping, brecciation of wallrock by intrusion of magma, intermittent explosive and/or implosive brecciation, spalling, slumping and possibly rock bursting, occur (FIGURE 5.2 A,B,C). Some of these processes may benefit from the presence of a precursor gas phase that evolves at the head of some kimberlite magmas. 133 The explosive breakthrough and eruption of kimberlite is predicted to occur within K300-400 m of the earth's surface. The kimberlite pipe results from the initial breach of the surface and the contemporaneous formation of explosion craters (Clement, 1982, FIGURE 5.2 D,E). Subsidiary implosion brecciation, referred to as authigenic breccias, generated by the release of pressure on pre-fragmented wallrock, allows the pipe to propagate to depths below the crater zone. The diatreme facies of kimberlite is ascribed to post-breakthrough modification of the basal part of the crater zone and considerable parts of the extended root zone. The diatreme zone develops as a vapour-liquid-solid fluidized system resulting from the rapid depressurization which accompanies explosive breaching of the surface (FIGURE 5.2 E). The result is a rapid upsurge of precursor gas phases and partly degassed magma. In other terms, the magma becomes saturated with C0 2 , which results in the nucleation, and expansion of bubbles until the melt is fragmented. Due to vaporization and rapid adiabatic expansion these systems are expected to evolve backwards down the embryonic pipes as the depressurization wave moves down through the magma column. The diatreme forming fluidization event is expected to have been short lived. Following the fluidization event, deflation and collapse of the magma column is followed by deposition of material derived from the erosion of tuff rings in the crater zone (e.g. epiclastic kimberlite, FIGURE 5.2 F). The main evidence used in support of this model includes: A.) The presence of abundant carbonate and serpentine in hypabyssal kimberlite suggests that the kimberlite magmas are extremely volatile rich. B.) The gradational transitions from uniform textured hypabyssal rock, to kimberlite characterized by calcite and serpentine segregations (segregationary textured kimberlite), to clast supported rocks dominated by small, generally nucleated, spheroids of kimberlite melt within a serpentine and clinopyroxene matrix where little to no carbonate is present. This transition is explained by the exsolution of C0 2 from the kimberlite magma (Clement, 1982; Clement and Reid, 1989). C.) The scarcity of primary calcite and low C 0 2 134 contents of diatreme facies rocks relative to hypabyssal kimberlite is thought to reflect the loss of C02 through degassing (Clement 1982, Skinner and Marsh 2003). Residual Ca and Mg are left from the loss of C02 from carbonate and these elements are thought to be the cause of clinopyroxene crystallization and serpentinization of olivine. D.) The presence of abundant and large downrafted blocks to depths in excess of 400m below their original position is considered to result from fluidization (Field et al. 1997) as is the entrainment and transport of large xenoliths to surface. Where flow velocities are high, blocks are entrained. Along the edges where velocities are lower, blocks are no longer suspended and fall under the force of gravity. E.) The uniformity of texture within diatreme facies is explained as resulting from turbulent mixing brought on by fluidization (Field et al. 1997). 5.2 The Phreatomagmatic model The phreatomagmatic model for kimberlite eruption is based predominantly on geometry and pyroclastic deposits of kimberlite and by analogy with other magma types. The kimberlites are modeled as diatremes and craters infilled with subsided pyroclastic deposits (Figure 5.3). Lorenz and Kurszlaukis (2003) equate kimberlites with maar volcanoes, which they believe form according to the following sequence (FIGURE 5.4, and 5.5). A rising magma intersects an aquifer at shallow depth during its ascent to surface. Entrainment of water or wet sediment into the magma results in molten fuel coolant interactions (MFCI) that result in violent explosions (FIGURE 5.5A). Thermohydraulic processes represent a very efficient mode of conversion of thermal energy into mechanical form and this process is possible only upon water entrapment inside the melt (Zimanowski et al., 1997). There are four phases to MFCIs : 1) Initial contact and coarse mixing of magma and water under stable film boiling conditions (Leidenfrost phenomenon). 135 2) Comple te vapor f i l m collapse cause either by the passage o f a pressure pulse which may be o f external origin or by a local implos ion due to rapid condensation o f coolant vapor. O n c e the coolant vapor is completely condensed, the fuel and coolant are thermally and mechanical ly coupled. 3) Ep i sod ic increase o f heat transfer from magma to water and fine fragmentation leading to superheated and pressurized water. A s the coolant is heated, it expands at a rate o f <1 u,s, leading to rapid increase in load stress on the melt. Relaxation o f load stress in the brittle mode (elastic rebound) causes the explosive release o f seismic energy. U p to 90% o f the total kinetic explosion energy is released in this phase. 4) Vo lumetr i c expansion o f the magma-water mixture from the transformation o f the superheated water to superheated steam. A t this stage, the melt and water are thermally and mechanical ly decoupled (Morrissey et al. 2000; Buttner and Z i m a n o w s k i , 1998). M F C I s fragment the magma as a direct result o f the explosion resulting in small angular 'active particles', and through the acceleration imparted from the explosion, w h i c h results in passive particles formed through shear thinning or decompressional growth. Passive particles can have f luidal shapes or spherical shapes. Initial explosions take place near the surface, resulting in a wide flat crater (F IGURE 5.3,5.4). T h e major products o f the explosions are base surges and pyroclastic fall deposits. W e t surges and accretionary pyroclasts are characteristic elements due to the high proportions o f water involved in phreatomagmatic eruptions. A s eruption proceeds the locus o f the eruption moves downward (FIGURE 5.4A and 5.5B,C) as the water table is depressed (F IGURE 5.4). Eruptions at depth create an open space, which is filled by subsidence o f the volcanic pile with continued eruption (F IGURE 5.4, and 5.5B-D). A s a result, the diatreme is f i l led with bedded deposits to considerable depths (FIGURE 5.3,5.4, & 5.5A-E). Unstructured material within the diatreme is material that has been disrupted due to later eruptions and so 136 primary bedding is now obscured. Phreatomagmatic eruption ends when either the water or magma supply ends (FIGURE 5.5 E, F). If the water supply ends first, then a transition to purely magmatic activity should ensue. If the magma supply ends, then all eruptive activity will cease. The main lines of evidence used to support the phreatomagmatic kimberlite eruption model are the similarity in geometry and deposit types between kimberlites and phreatomagmatic deposits of other magma types (Lorenz, 1991) and the similarities between clasts observed from kimberlite and those observed from experimental phreatomagmatic explosions (Lorenz et al. 1991), Zimanowski (1998). It should be noted that the above eruptive model specifically refers to MFCIs as the mechanism for explosions; however; other processes such as the explosive expansion and collapse of steam formed at magma-water contact surfaces and cooling contraction are also included in the range of possible phreatomagmatic interactions. Field and Scott Smith (1999) propose a model for kimberlites based on investigation of pipes in the prairie regions of Canada. This model is a variation on the phreatomagmatic model detailed above. In this model kimberlite craters are completely excavated, through phreatomagmatic explosions, and are subsequently infilled by pyroclastic kimberlite through Hawaiian style fire-fountaining. 5.3 Limitations of current models One of the main criticisms of the MFCI-phreatomagmatic model is that it excludes the possibility of exsolution driven fragmentation. MFCIs are prevented if the vesiculation is more than ~ 15 vol% (Zimanowski, 1998). In contrast exsolution driven fragmentation requires volatile contents closer to ~75% (Sparks, 1978) for fragmentation. This is problematic because even magmas with relatively low volatile contents will reach saturation in the near surface unless 137 they are able to degas effectively and kimberlites are typically associated with high volatile solubilities. Detailed examination of South African kimberlites also (Clement 1982) does not support the phreatomagmatic model of a crater infilled by bedded deposits. A second criticism of the phreatomagmatic model is that no non-explosive kimberlites have been documented, with the possible exception of Igwisi Hills (Mitchell 1986). It seems unlikely that all erupted kimberlites would have interacted phreatomagmatically, and more likely that these volcanoes would have erupted regardless of the presence of water. A final note, is that a very specific set of conditions are necessary in order to generate MFCIs. These include specific differential flow conditions, viscosities, time scales, triggering and pressure temperature conditions. As a result, it seems reasonable that although MFCIs may contribute to kimberlite eruption, they are not likely to play a dominant role, and should be not expected at all kimberlite volcanoes. However phreatomagmatism in general (including MFCI's, quench fragmentation, etc.) is likely an important enhancement of primarily magmatic kimberlite eruptions. The exsolution model alone cannot account for the features observed in BHH kimberlites. The abundant accretionary pyroclasts and common surge deposits are strong indications that phreatomagmatism was an important eruptive mechanism, and as a result exsolution driven volcanism alone, does not explain these features. If volatile-rich basalts are considered as an analogue for kimberlite volcanism, the deposits expected would resemble spatter cones, with abundant coarse juvenile material. BHH kimberlites have a finer clast size population than typical spatter cones, most of the juvenile pyroclasts appear to have quenched prior to deposition and generally do not have features consistent with Hawaiian style fountaining, or other exsolution driven magmatism. The presence of accretionary pyroclasts and surge deposits indicatees that phreatomagmatism, but not necessarily MFCI, was an important mechanism in this region. The possibility of exsolution 138 induced fragmentation is however not eliminated, the previous facts simply indicate that it was not the sole mechanism responsible for the features observed in BHH kimberlites. Other objections to the exsolution model include the similarity of extra-crater and crater kimberlite deposits to maar-style volcanoes. This similarity includes the presence of surge deposits, fall deposits, accretionary pyroclasts, and geometries that are generally consistent with maars. These features are used to infer the likelihood of phreatomagmatic eruption of kimberlite. 5.4 Eruptive model for Buffalo Head Hills Kimberlite Neither of the two end-member models, reviewed above, fully explain the deposits found in BHH kimberlites. Here, I present an integrated volcanological model for the eruption of BHH kimberlites. This model differs from previous models in the integration of both exsolution driven and phreatomagmatic volcanism. The model has the following basic elements: a. ) Initiation of ascent of a kimberlite magma with an initial volatile content in excess of 1 wt% C 0 2 + H 20. Ascent is initiated by the result of buoyancy contrast between the melt and surrounding rock (Spera, 1984, Lister and Kerr 1991). b. ) Decreasing pressure associated with magma ascent results in saturation of the magma with respect to volatile species and after a certain overpressure nucleation of bubbles will be initiated (Mader, 1998) (FIGURE 4.6A) c. ) Once nucleated, bubbles begin to grow by diffusion of volatiles from the melt and by expansion due to decompression (Sparks, 1978). Progressive bubble growth causes the mixture to expand and results in an upwardly accelerating flow (FIGURE 4.6A). If the overall velocity of the magma is high, as modeled by Spera (1984), McGetchin and Ullrich (1973), Anderson (1979) and Wilson and Head (2003), then bubble coalescence is negligible and the possibility of separated flow is unlikely (Parfitt, 2004). 139 d. ) At some height in the conduit, the liquid is fragmented into discrete particles and thereafter the flow is of a gas-particle mixture ( F I G U R E 4.6A). If initial volatile contents are high, on the order of 12 wt% (Kjarsgaard, 2003) then the system will meet the condition of 70-80% volatiles by volume, which is the criteria for magma fragmentation (Sparks 1978) at depths of ~ 2km. If the volatile content is lower, fragmentation will occur closer to surface. For example 5wt% initial would be expected to result in fragmentation at approximately 500 m below surface (Kjarsgaard, 2003). If the volatile content is higher fragmentation will be expected a greater depths. This fragmentation represents a transition from volatile domains dispersed in magma, to magma domains dispersed in the volatile phase (Cashman et al. 2000). e. ) Intrusion into wet sediments results in the incorporation of external water into the gas particle dispersal. Turbulent mixing results in efficient heat transfer from the hot intrusive mixture and the newly introduced water and sediment. The rheological properties of the kimberlite, including xenolith content, degree of crystallinity, and volatile content, will dictate whether the kimberlite flows in laminar regime or a turbulent regime. If the kimberlite is flowing in a laminar fashion and is a non-dispersed flow, there is potential for MFCIs in the near surface. If the kimberlite is flowing turbulently, either dispersed or not, phreatomagmatic interaction is still likely as the kimberlite incorporates water or wet sediments ( F I G U R E 4.6A), however the mechanisms will be more akin to quench fragmentation and steam expansion (Mastin, 2004). The turbulent mixing, results in quenching of pyroclasts and the generation of steam. This mechanism creates a feedback loop that increases ascent rate and turbulence. f. ) The surface is breached and the eruption is sustained by the steady removal of material from the volcano, which causes the fragmentation surface to propagate downwards in response to off-loading and decrease in pressure ( F I G U R E 4.6B; Clement, 1982). Initially 140 • • ' 1 1 • Figure 5.6: Schematic representation of proposed kimberlite eruptive sequence. A. Intrusion of kimberlite magma. Ascending magma becomes saturated with volatiles and bubbles begin to nucleate. Bubbles then grow through diffusion and decompression until a critical volume percent of volatiles is reached, and the melt fragments. During ascent, water and wet sediment is entrained from surrounding Cretaceous sediments. B. The magma breaches the surface and excavates a crater. Pyroclastic surges and a volcanic column are produced. The fragmentation level migrates downwards due to depressurization associated with surface breach. C. Pyroclastic surge and fall results in the construction of a tephra ring. As eruptive activity subsides, column collapse feeds subsequent surges and fills the crater. D.The morphology of the kimberlite after eruptive activity has ceased.The tephra ring is dominated by surge deposits and the crater consists dominantly of fall material. E. Erosion results in flattening of the tephra ring and infill of the crater depression with reworked volcaniclastic deposits. Abbreviations: RVK=reworked volcaniclastic kimberlite, PS=pyroclastic surge, PF=pyroclastic fall. 141 shallow explosions will result in the formation of a wide shallow crater, but as the fragmentation surface propagates downwards, deeper and deeper explosions will result in an increasingly narrow chimney or diatreme (Rice, 1999). g. ) The continuous to semi-continuous flow of material from the vent results in the formation of a volcanic plume. Pyroclastic surge deposits are initiated as a result of condensation of water and subsequent partial column collapse resulting in dilute density currents (FIGURE 4.6B). A volcanic column is set up and low concentration density currents (base surges) are initiated that move radially outward from the vent. Fluctuations in column dynamics result in partial column collapses that feed surges. Tuff ring deposits are built up through aggradational sedimentation from surges as well as pyroclastic fall. h. ) Continued eruption results in the formation of a tephra ring dominated by pyroclastic surge and fall deposits and a crater infilled by proximal fall and slumped material (FIGURE 4.6c). Proximal pyroclastic fall and synvolcanic slumping of material from the crater rim partially infill the crater, final crater infill results from collapse of the pyroclastic column (F IGURE 4.6C,D). i. ) Erosion of tephra ring and deposition into crater acts to flatten topography (FIGURE 4.6E) and glaciation removes material. Typically, exsolution-driven and phreatomagmatic models for kimberlite eruption have been considered mutually exclusive mechanisms. This results from the assertion that MFCIs are the mechanism for phreatomagmatic eruption of kimberlite. However, MFCIs are essentially impossible at volatile contents of 15 vol%. However, MFCIs are not the only mechanisms whereby a magma can erupt phreatomagmatically. A combination of phreatomagmatism and exsolution driven magmatism is proposed to explain the eruptive mechanisms of BHH kimberlites. This combination does not adhere 142 completely to any previously proposed model for kimberlite volcanism but does incorporate aspects of previous models. The mechanism envisaged is similar to that proposed by Mastin et al. (2004) for the Keanakako'i ash, from Kilauea volcano Hawaii. In this model, high effusion rates of magma result in turbulent flow that allows mixing with surrounding wet sediments and any water present. The result is a rapid heat exchange leading to extensive magma fragmentation and quenching of pyroclasts. The driving mechanisms for magma ascent would result from buoyancy of the melt and possible exsolution of volatile phases such as C0 2 and H 20. The interaction with external water would intensify the violence of these eruptions, but is not considered the driving force for the eruption. The model proposed accounts for the observed phreatomagmatic features as well as allowing for significant volatile contents to be present within the melt. There are several reasons why this model is preferred for the eruption of kimberlite. One of the first considerations is that known kimberlites have violent eruption histories. Quiet effusive events are rare to non-existent. As a result, it seems reasonable to conclude that kimberlite erupts violently regardless of the hydrogeologic regime into which it intrudes. Kimberlites are also associated with intense carbonate and serpentine alteration in virtually all localities. This alteration reflects initially high volatile contents in kimberlite, as do estimates from geochemical analyses of hypabyssal kimberlite (Price et al. 2000). Models of solubility of C 0 2 and H 20 in kimberlite indicate that they can be present in very large amounts (Brey and Ryabchikov,1994; Brey et al., 1991; Wyllie, 1979; and Spera, 1984) One argument against this model is that exsolution driven magmatism typically results in vesicular juvenile pyroclasts, as exsolution occurs in a continuum where the melt is constantly exsolving volatiles. The pyroclasts observed from BHH kimberlites are dominantly non-vesicular, with only rare vesicles observed. Explanations for the lack of vesicularity can be presented, such as the low viscosity of kimberlite liquid permitting easy escape of volatiles, the 143 rapid micro-crystallization of kimberlite melt, and pervasive alteration of kimberlite samples. Conversely however, it is difficult to envision a melt that was exsolving no volatile species and contained no vesicles. Estimates of C 0 2 quantities in hypabyssal kimberlite from worldwide localities indicate relatively high quantities (Price et al. 2000), and this would also argue for exsolution of volatiles occurring at some point. 144 Chapter 6: Conclusion Examination of six of the Buffalo Head Hills kimberlite occurrences provides unique insights into the processes of kimberlite eruption. The paleo-environment at the time of kimberlite eruption is ideal for phreatomagmatism and many of the aspects of BHH kimberlites are consistent with an eruptive style that is distinctly phreatomagmatic. Pyroclastic surge and fall deposits dominate the volcanic facies observed, and accretionary pyroclasts are common; strong indicators of phreatomagmatism. However, previous studies of kimberlite indicate a large role for magmatically exsolved volatiles in kimberlite fragmentation and eruption. Kimberlites in the BHH appear to be no exception. Intense serpentinization and carbonatization of kimberlite is pervasive, with preservation of primary mineralogy the exception rather than the norm. These processes imply the presence of C0 2 , which is supported by high estimates for volatile solubility in kimberlite. Accordingly, these two processes are considered to work in concert in BHH kimberlites. Fragmentation of the kimberlite magma through exsolution driven processes, increases ascent rate and promotes turbulent mixing with surrounding sediments and water. Heat transfer between the magmatic mixture and external water sediment mixture results in the quenching of magmatic clasts and the conversion of meteoric water to steam, which consequently acts to further increase ascent rate. These processes result in what is beneath' the surface of the BHH today, kimberlite volcanoes that consist of a large central crater surrounded by a tephra ring. The crater is infilled first by pyroclastic fall and synvolcanic slumping of the tephra ring, and later by resedimented kimberlite derived from the tephra ring. The tephra ring is dominated by pyroclastic surge and fall deposits, and thins rapidly with distance from the vent. Outside of the tephra ring, pyroclastic 145 fall deposits and reworked pyroclastic kimberlite are found as the most distal records of kimberlite volcanism. 146 References Anderson, O.L., 1979. The role of fracture dynamics in kimberlite pipe formation. In: Boyd, F.R., and Meyer, H.O.A., (eds) Kimberlites, Diatremes, and Diamonds. Their Geology, Petrology and Geochemistry. Proceedings of the 2nd International Kimberlite Conference. Vol.1, American Geophysical Union, Washington, D.C., p. 344-353 Ashton Mining of Canada Inc.,1997. Ashton discovers kimberlitic pipes in Alberta. Press Release dated January 27,1997, p.2. Ashton Mining of Canada Inc., 2003a. Second Kimberlite Discovered during Alberta program. Press Release dated February 27, 2003, p.l Ashton Mining of Canada Inc., 2003b. Alberta K296 Kimberlite Diamondiferous. Press Release dated May 16, 2003 p.l Ashton Mining of Canada Inc., 2001. K252 Mini-Bulk Sample Returns 55.0 CPHT. Press Release dated May 24, 2001 p. 1 Ashton Mining of Canada Inc., 2000. Preliminary Internal Report on Work: The Geologic Setting of Alberta Kimberlites, Geochronological, Palynological and Petrographic Evidence. Unpublished. Bhattacharya, J., and Walker, R.G., 1991. Allostratigraphic subdivision of the Upper Cretaceous Dunvegan, Shaftesbury, and Kaskapau formations in the northwestern Alberta subsurface. Bulletin of Canadian Petroleum Geology, Vol. 39, No.2, p. 145-164 147 Brey, G.P., Kogarko, L.N., and Ryabchikov, 1991. Carbon dioxide in kimberlitic melts. N. Jb. Miner. Mh. jg.1999, H.4, p. 159-168 Brey, G.P. and Ryabchikov, I.D., 1994. Carbon dioxide in strongly silica undersaturated melts an origin of kimberlite magmas. N. Jb. Miner. Mh., H. 10, p. 449-463 Burwash, R.A., Krupicka, J. and Wijbrans, J.R., 2000. Metamorphic evolution of the Precambrian basement of Alberta. Canadian Mineralogist 38. p. 423-434, Buttner, R., and Zimanowski, B., (1998) Physics of thermohydraulic explosions. Physical Review E. Volume 57, Number 5. p. 5726-5729 Carlson, S.M., Hillier, W.D., Hood, C.T., Pryde, R.P., and Skelton, D.N., 1999. The Buffalo Hills Kimberlites: A Newly-Discovered Diamondiferous Kimberlite Province in North-Central Alberta, Canada. In: Gurney, J.J., Gurney, J.L., Pascoe, M.D., and Richardson, S.H., (eds) Proceedings of the VHth International Kimberlite Conference, The J.B. Dawson Volume, p.109-116 Cas, R.A.F. and Wright, J.V.,1987. Volcanic Successions: Modem and Ancient. Allen & Unwin, London. Cashman, K.V, Sturtevant, B., Papale, P., and Navon, O., 2000. Magmatic Fragmentation In: Sigurdsson, H.,(ed) Encyclopedia of Volcanoes. Academic Press, p. 421-430 Clement, C.R., 1982. A comparative geological study of some major kimberlite pipes in the Northern Cape and Orange Free State, Unpublished. Ph.D. Thesis. Univeristy of Cape Town. 2 vols. Clement, C.R., and Reid, A.M., 1989. The origin of kimberlite pipes: An interpretation based on a synthesis of geological features displayed by southern African 148 occurrences. Geological Society of Australia Special Publication., 14, 1, p. 632-646 Clement, C.R. and Skinner,E.M.W., 1979. A textural-genetic classification of kimberlite rocks. Kimberlite Symposium n , Cambridge, Extended Abstract. Clement, C.R. and Skinner, E.M.W., 1985. A textural-genetic classification of kimberlites. Trans. Geol. Soc. S.Afr. 88, p. 403-409. Core laboratories, Petroleum Services. Core data sheet. Calgary, Alberta Dawson, J.B. 1980. Kimberlites and their xenoliths. Springer-Verlag, Berlin, Heidelberg, New York. Dellino, P., Isaia, R., La Volpe, L., and Orsi, G., (2001) Statistical analysis of textural data from complex pyroclastic sequences: implications for fragmentation proceses of the Agnano-Monte Spina Tephra (4.1 ka), Phlegraean Fields, southern Italy. Bulletin of Volcanology 63 No.7: 443-461 Dufresne, M.B., Eccles, D.R., and Leckie, D.A., (2001) The Geological and Geochemical Setting of the Mid-Cretaceous Shaftesbury Formation and Other Colorado Group Sedimentary Units. Alberta Geological Survey Special Report 009. Edwards, C.B. and Howkins, J.B., 1966. Kimberlites in Tanganyika with special reference to the Mwadui occurrence. Economic Geology, 61. p. 537-554 Ernst, G.G.J., Durant, A.J., Rose, W.I., and Self, S., (2004) Accretionary lapilli and water-rich eruption columns: new data and insights. In: Nemeth, K., Martin, U., Goth, K., and Lexa, J., (eds). Occasional Papers of the Geological Institute of Hungary, volume 203. Abstract Volume of the Second International Maar Conference. 149 Field, M., Gibson, J.G., Wilkes, T.A., Gababotse, J., and Khutjwe, P., (1997) The Geology of the Orapa A/Kl Kimberlite Botswana: Further Insight into the Emplacement of Kimberlite Pipes. Russian Geology and Geophysics. Vol 38. No.l p, 24-39 Field, M., and Scott Smith, B.H., (1999) Contrasting Geology and Near-Surface Emplacement of Kimberlite Pipes in Southern Africa and Canada. Proceedings of the Vllth International Kimberlite Conference. The J.B. Dawson Volume. J.J. Gurney, J.L. Gurney, M.D. Pascoe and s.H. Richardson. Volume 1. p.214-237 Fisher, R.V., and Schmincke, H.U., (1984) Pyroclastic Rocks, Springer Verlag Gilbert, J.S. and Lane, S.J., 1994. The origin of accretionary lapilli. Bulletin of Volcanology 56 (5). p. 398-411 and Lane, 1994 Bulletion of Volcanology Hamilton, W.H., Langenberg, W., Price, M.C., and Chao, D.K., 1998. Map 236: Geological map of Alberta. Geological Survey of Alberta Hawthorne, J.B., (1975) Model of a Kimberlite Pipe. Physicas and Chemistry of the Earth, 9, p.1-15 Hetman, C M . , Scott Smith, B.H., Paul, J.L. and Winter, F.W., 2003. Geology of the Gahcho Kue Kimberlite Pipes, NWT, Canada: Root to Diatreme Transition Zones. In: Extended Abstracts of the 8th International Kimberlite Conference. Kjaarsgard, B.A., 2003. Behaviour of Kimberlite Magma in the Upper Crust and at Surface. In: Extended Abstracts of the 8th International Kimberlite Conference. Kurszlaukis, S., Buttner, R., Zimanowski, B., and Lorenz, V., (1998) On the first experimental phreatomagmatic explosion of a kimberlite melt. Journal of volcanology and geothermal research 80: 323-326 150 Leahy, K., 1997. Discrimination of reworked pyroclastics from primary tephra-fall tuffs: a case study using kimberlites of Fort a la Come, Saskatchewan, Canada. Bulletin of Volcanology vol 59, p. 65-71 Lister, J.R., and Kerr, R.C. 1991. Fluid-mechanical models of crack propagation and their application to magma transport in dykes. Journal of Geophysical Research, 96; B6. p. 10049-10077 Lorenz, V. 1986. On the growth of maars and diatremes and its relevance to the formation of tuff rings. Bulletin of Volcanology, 48, p. 265-274. Lorenz, V. 1987. Phreatomagmatism and its relevance. Chemical Geology 62 p. 149-156 Lorenz, V., Zimanowski, B., and Frohlich, G., 1991. Experiments on Explosive Basic and Ultrabasic, Ultramafic and Carbonatitic Volcanism. In: Kimberlites, Related Rocks and Mantle Xenoliths, Proceedings of the Fifth International Kimberlite Conference. Volume 1. p.270-284 Lorenz, V. and Kurszlaukis, S., (2003) Kimberlite pipes: Growth Models and Resulting Implications for Diamond Exploration. In: Extended Abstracts of the 8th International Kimberlite Conference. Mader, H.M., 1998. Conduit flow and fragmentation In: Gilbert, J.S., & Sparks, R.S.J.,(eds) The Physics of Explosive Volcanic Eruptions. Geological Society, London, Special Publications, 145, 51-72. McGetchin, T.R., and Ullrich, W.U., (1973) Xenoliths in maars and diatremes with inferences for the Moon, Mars, and Venus. Journal of Geophysical Research, 78;11: p. 1833-1853 151 Mastin, L.G., (2004) The hydrodynamics of magma-water mixing and its effect on fragmentation and eruptive violence. In: Abstract Volume of the Second International Maar Conference. Occasional Papers of the Geological Institute of Hungary, volume 203. p.76 Mastin, L.G., Christiansen, R.L., Thornber, C , Lowenstern, J. and Beeson, M. 2004. What makes hydromagmatic eruptions violent? Some insights from the Keanakako'i Ash, Kilauea Volcano, Hawai'i. Journal of Volcanology and Geothermal Research 137. p. 15-31. Mannard, G. 1962. The Singida kimberlite pipes. Unpublished Ph.D. thesis, McGill University. Mitchell, R.H., (1986) Kimberlites. Plenum Press, New York and London. Mitchell, R.H., (1995) Kimberlites, Orangeites, and Related Rocks. Plenum Press, New York and London. Mitchell, R.H., and Bergman, S.C., (1991) Petrology of Lamproites. Plenum Press, New York. Morrissey, M., Zimanowski, Bernd., Wohletz, K., and Buettner, R., (2000) Phreatomagmatic Fragmentation. In: Sigurdsson, H. (ed) Encyclopedia of Volcanoes, Academic Press, p. 431-444 Mossop G.D., and Shetson, I. (compilers) 1994. Geological Atlas of the Western Canada Sedimentary Basin. Canadian Society of Petroleum Geologists and Alberta Research Council, Calgary. 152 Parfitt, E.A., (1998) A study of clast size distribution, ash deposition and fragmentation in a Hawaiian-style volcanic eruption. Journal of Volcanology and Geothermal Research 84. p. 197-208 Parfitt, E.A., (2004) A discussion of the mechanisms of explosive basaltic eruptions. Journal of Volcanology and Geothermal Research 134. p.77-107 Pawlowicz, J.J., and Fenton, M.M., (1995) Drift thickness of Alberta. Alberta Geologic Survey Map. Price, S.E., Russell, J.K., and Kopylova, M.G., 2000. Primitive Magma from the Jericho Pipe, N.W.T., Canada: Constraints on Primary Kimberlite Melt Chemistry. Journal of Petrology 41, 6, p 789-808 QC Data Ltd., (1999). Basemap & Well Summary Rice, A., 1999, Can the Blasting and Excavation Engineering Sciences Provide Insight into the Processes of Kimberlite Emplacement and Eruption? In: Gurney, J.J., Gurney, J.L., Pascoe, M.D., and Richardson, S.H. (eds) Proceedings of the VHth International Kimberlite Conference, P.H. Nixon Volume, p.699-708 Ross, G.M. 2002. Evolution of Precambrian continental lithosphere in Western Canada: results from Lithoprobe studies in northern Alberta. Canadian Journal of Earth Sciences 39 (3) p. 413-437. Ross, G.M., and Eaton, D.W., 1997. Winagami reflection sequence: seismic evidence for post-collisional magmatism in the Proterozoic of western Canada. Geology 25. p. 199-202 153 Ross, G.M., Parrish, R.R., Villeneuve, M.E., and Bowring, S.A., 1991. Geophysics and geochronology of the crystalline basement of the Alberta Basin, western Canada. Canadian Journal of Earth Sciences 28. p. 512-522 Ross, G.M. (1990) Deep crust and basement structure of the Peace River Arch region: constraints on mechanisms of formation. Bulletin of Canadian Petroleum Geology Special Volume 38A: 25-35 Scott Smith, B.H., (1992). Contrasting kimberlites and lamproites. Explor. Mining Geol. 1, p. 371-382 Skelton, D.N., Clements, B., McCandless, T.E., Hood, C , Aulbach, S., Davies, R., and Boyer, L.P., (2003) The Buffalo Head Hills Kimberlite Province, Alberta. In: GuideBook of the 8ikc. Skinner, E.M.W., and Marsh, J.S. (2003) Kimberlite Eruption Processes. In: Extended Abstracts of the 8lh International Kimberlite Conference. Skinner, E.M.W., and Clement, C.R., (1979) Mineralogical classification of southern African kimberlites. In Boyd and Meyer (1979) q.v., Vol.1, p. 129-139 Sparks, R.S.J., Bursik, M.I., Carey, S.N., Gilbert, J.S., Glaze, L.S., Sigurdsson, H., and Woods, W., (1997) Volcanic Plumes. Wiley Sparks, R.S.J., (1978). Dynamics of bubble formation and growth in magmas - Review and analysis. Journal of Volcanology and Geothermal Research 3 (1-2). p. 1-37 Spera, F.J., (1984) Carbon dioxide in petrogenesis III: role of volatiles in the ascent of alkaline magma with special reference to xenolith-bearing mafic lavas. Contributions to Mineralogy and Petrology. Volume 88. p. 217-232. 154 Theriault, R.J. and Ross, G.M., (1991). ND isotopic evidence for crustal recycling in the ca. 2.0 GA subsurface of western Canada. Canadian Journal of Earth Sciences 28. p. 1140-1147 Tremblay, M. 1956. Geology of the Williamson Diamond Mine, Mwadui, Tanganyika. Unpublished Ph.D. thesis, McGill University. Valentine, G. A., and Fisher, R.V., (2000) Pyroclastic surges and blasts. In: Sigurdsson, H. (ed) Encyclopedia of Volcanoes, Academic Press, p. 571-580 Webb, K.J., Stiefenhofer, J. and Field, M. 2003. Overview of the Geology and Emplacement of the Jwaneng DK2 Kimberlite, Southern Botswana. In: Extended Abstracts of the 8th International Kimberlite Conference. Wall, J.H. and Singh, C. 1975. A Late Cretaceous microfossil assemblage from the Buffalo Head Hills, north-central Alberta. Canadian Journal of Earth Sciences, v. 12, p. 1157-1174 White, J.D.L., (1996). Impure coolants and interaction dynamics of phreatomagmatic eruptions. Journal of Volcanology and Geothermal Research 74. pp. 155-170 Wilson, L. and Head, J.W. 2003. Diatremes and Kimberlites 2: An Integrated Model of the Ascent and Eruption of kimberlitic magmas and the producation of crater, diatreme and hypabyssal facies. In: Extended Abstracts of the 8th International Kimberlite Conference. Wyllie, P.J., 1979. Magmas and volatile components. American Mineralogist 64, p.469-500. 155 Zimanowski, B., (1998) Phreatomagmatic Explosions. In: Freundt, A. and Rosi, M. (eds) Developments in Volcanology 4. From Magma to Tephra, modeling phyrsical processes of explosive volcanic eruptions. Elsevier p.25-54 Zimanowski, B., Buttner, R., Lorenz, V. and Hafele, H.G., 1997. Fragmentation of basaltic melt in the course of explosive volcanism. Journal of Geophysical Research-Solid Earth 102. p.803-814 et al. 156 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items