A practical guide to terminology for kimberlite facies: A systematic progression from descriptive to genetic, including a pocket guide

Lithos 112S (2009) 183–190

Contents lists available at ScienceDirect

Lithos j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / l i t h o s

A practical guide to terminology for kimberlite facies: A systematic progression from descriptive to genetic, including a pocket guide R.A.F. Cas ⁎, L. Porritt, A. Pittari 1, P.C. Hayman School of Geosciences, Monash University, Clayton, Victoria, 3800 Australia

a r t i c l e

i n f o

Article history: Received 1 October 2008 Accepted 28 March 2009 Available online 23 April 2009 Keywords: Descriptive to genetic terminology Coherent kimberlite Fragmental kimberlite Alteration effects Practical approach

a b s t r a c t Kimberlite terminology remains problematic because both descriptive and genetic terms are mixed together in most existing terminology schemes. In addition, many terms used in existing kimberlite terminology schemes are not used in mainstream volcanology, even though kimberlite bodies are commonly the remains of kimberlite volcanic vents and edifices. We build on our own recently published approach to kimberlite facies terminology, involving a systematic progression from descriptive to genetic. The scheme can be used for both coherent kimberlite (i.e. kimberlite that was emplaced without undergoing any fragmentation processes and therefore preserving coherent igneous textures) and fragmental kimberlites. The approach involves documentation of components, textures and assessing the degree and effects of alteration on both components and original emplacement textures. This allows a purely descriptive composite component, textural and compositional petrological rock or deposit name to be constructed first, free of any biases about emplacement setting and processes. Then important facies features such as depositional structures, contact relationships and setting are assessed, leading to a composite descriptive and genetic name for the facies or rock unit that summarises key descriptive characteristics, emplacement processes and setting. Flow charts summarising the key steps in developing a progressive descriptive to genetic terminology are provided for both coherent and fragmental facies/ deposits/rock units. These can be copied and used in the field, or in conjunction with field (e.g. drill core observations) and petrographic data. Because the approach depends heavily on field scale observations, characteristics and process interpretations, only the first descriptive part is appropriate where only petrographic observations are being made. Where field scale observations are available the progression from developing descriptive to interpretative terminology can be used, especially where some petrographic data also becomes available. © 2009 Elsevier B.V. All rights reserved.

1. Introduction The terminology used in kimberlite geology contains many terms that are not used in volcanology, even though kimberlite bodies are often the remains of volcanic conduits, vents and edifices. The existing terminology consists of a mixture of descriptive and genetic terms, and terms that are not used in mainstream volcanology, making it difficult to understand kimberlite terminology by geologists not familiar with its pecularities. Furthermore, some terms and concepts are ill-defined, which then confuses understanding of kimberlite geology. To overcome these weaknesses Cas et al. (2008a) proposed a systematic

⁎ Corresponding author. Tel.: +61 3 99054897; fax: +61 3 99054903. E-mail addresses: [email protected] (R.A.F. Cas), [email protected] (L. Porritt), [email protected] (A. Pittari), [email protected] (P.C. Hayman). 1 Present address: Department of Earth and Ocean Sciences, University of Waikato, Private Bag 3105, Hamilton, 3240 New Zealand. 0024-4937/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.lithos.2009.03.051

approach to kimberlite facies terminology involving the progressive, systematic development of terminology from descriptive to genetic, based on the approaches cultivated by Cas and Wright (1987) and McPhie et al. (1993) for volcanic successions of all compositions. In this paper we outline some of the weaknesses of existing kimberlite terminology, we summarise the key steps in developing descriptive and then genetic terminology, we provide an update on the Cas et al. (2008a) paper, and we provide a compact, ready to use guide in the form of two reference charts. These represent a compilation of many of the tables in the Cas et al. (2008a) paper in a compact “pocket guide” that can be printed, and then used in the field based on field data alone, or combined with laboratory petrographic data. Several tables have been revised from the Cas et al. (2008a) paper. The approach advocated has been tested for several years by the authors on many different volcanic successions, including both kimberlite and non-kimberlite successions. It is thus well trialled and has been shown to be applicable in all types of volcanic successions and settings, including kimberlites.

184

R.A.F. Cas et al. / Lithos 112S (2009) 183–190

2. Weaknesses with existing kimberlite terminology schemes The major weaknesses and limitations associated with existing terminological approaches used in kimberlite geology are: 1. Some schemes are largely mineralogical only (e.g. Skinner and Clement, 1979; Smith, 1983; Mitchell, 1995; Woolley et al., 1996), and although very useful petrologically, they have limitations when considering deposit or facies characteristics and the volcanological origins. These schemes also do not adequately distinguish between primary minerals and secondary or alteration minerals. 2. Schemes that are more textural in their approach (e.g. Clement, 1982; Clement and Skinner, 1985; Mitchell, 1986, 1995; Field and Scott Smith, 1998) also lack the larger scale facies character considerations, which inform the process origins. In addition, those schemes ignore the extremely important effects of alteration on original depositional textures (Sparks et al., this issue; Hayman et al., this issue). In our experience of examining some 20 kimberlite bodies in detail, alteration affects all kimberlites (and lamproites) to varying degrees. Unless alteration effects are back-stripped to identify relict original emplacement textures, interpretation of the process origins of deposits is likely to be misleading. 3. Most kimberlite terminology schemes have lacked input from physical volcanologists, and as a result, some kimberlite terms that have been applied historically in kimberlite geology are not used at all, or are only rarely used, in mainstream volcanology (e.g. hypabyssal, autolith, tuffisitic kimberlite and tuffisitic kimberlite breccia, magmatic (as used in kimberlite geology), magmaclast, and magmaclastic). Such terms are variously unclear, ambiguous and confusing in their meaning, and now have historical significance only. For the sake of the advancing understanding of kimberlite geology in the context of well-understood volcanological processes, it is recommended that these historical terms be discarded and replaced by terminology that is explicitly descriptive, and does not require interpretation. Cas et al. (2008a) have outlined in detail the reasons why these terms should be discarded, and in some cases proposed appropriate alternative replacement terms. 4. With most kimberlite terminology schemes there is too much focus on the expected location of a deposit type or facies within a standard model kimberlite pipe, based on the southern African pipe model (e.g. Mitchell, 1986; Field and Scott Smith, 1999). The “model” kimberlite pipe has three morphological zones (crater zone, diatreme zone, root zone) and there is an inherent prediction and expectation as to what types of facies should be found in different parts of a kimberlite pipe. As outlined by Field and Scott Smith (1999) and Cas et al. (2008b), however, not all kimberlite pipes are “model Southern African” shaped pipes, as discovered when many Canadian kimberlite bodies were explored. For example, in the Koala kimberlite pipe of the Ekati kimberlite field in northwestern Canada, the crater zone is not preserved, but the diatreme and root zone are. However, crater zone facies (resedimented, pyroclastic) occur in the diatreme part of the pipe (e.g. Nowicki et al., 2004), overlying more typical, massive, poorly sorted lapilli-tuff “diatreme zone facies”. Interestingly, this tells us that the diatreme was in large part empty or unfilled at the end of the eruption that produced the Koala pipe, which is not an expectation of the traditional kimberlite model. Many of the Fort a la Corne, prairies style kimberlite bodies of central Canada are not deep, steep-sided tapering pipes, but the remains of shallow, open flaring craters, and some are the remains of positive relief supra-crustal cone-like edifices (e.g. Zonneveld et al., 2004; Pittari et al., 2008). This was not known until a complete three-dimensional facies architecture reconstruction was undertaken of some bodies based on detailed drill core logging and correlation. The shape of this style of kimberlite body is similar to that of some lamproite bodies (e.g. Smith and Lorenz, 1986), and it appears that the factors that

influence the shape of kimberlite and lamproite conduits and vents are common, as they are for all volcano types (see Cas et al., 2008b for discussion of these factors). Including consideration of the setting of a particular deposit or facies early in the process of documenting a facies and developing a facies name is therefore a huge mistake, especially in old volcanic successions and settings where erosion and even tectonic effects could have masked the original setting. We must be prepared for the unexpected and have a scheme for nomenclature that is purely descriptive and based on the deposit characteristics in the first stages, otherwise genetic preconceptions interfere with our inclinations in deciding on the appropriate terminology. Consideration of the setting of a facies should therefore be one of the last steps or descriptors included in a name, not the first. This cannot be stressed strongly enough. Field and Scott Smith (1998) proposed what can be described as a hybrid descriptive–genetic scheme in that it uses a combination of terms, some of which are descriptive, and others genetic, but not in a progressive order from descriptive to genetic. As discussed above some terms and their usage in their scheme are misleading (e.g. magmaclast), and others are unclear in their meaning or significance (e.g. tuffisite/tuffisitic kimberlite, hypabyssal). Sparks et al. (2006) were the first to publish their concerns about the traditional terminology used in kimberlite geology, and made some valuable suggestions for better, descriptive based terms. They did however adopt much of the approach and terminology of Field and Scott Smith (1998), and did not propose a comprehensive and systematic approach to terminology. Based on descriptive deposit characteristics, they proposed that tuffisitic kimberlite should be called massive volcaniclastic kimberlite, but accepted that hypabyssal kimberlite could be called magmatic kimberlite, which is a tautology because all kimberlite is magmatic in origin. Scott Smith et al. (2008) proposed a nomenclature scheme that is an attempt to distinguish descriptive from interpretative terminology, and is extremely similar in general approach to the scheme published by Cas et al. (2008a). However, the scheme lacks the detailed guidelines and steps for developing logically a progression from clearly descriptive terminology to genetic terminology, it confuses descriptive and genetic terms in places, it has difficulties in replacing some problematic terms of the old terminology such as “tuffisite/ tuffisitic”. As suggested by them this could be called Wesselton-type volcaniclastic kimberlite, a term that means nothing to anyone who has not seen the Wesselton pipe deposits, and is not constructed through logical guidelines based on the descriptive characteristics of the deposits. They also propose that consideration of setting should be one of the first steps in classification, which as noted above, is potentially a major mistake. 3. The approach to kimberlite terminology The rationale behind the approach to developing terminology involving a progression from descriptive to genetic has been outlined in detail by Cas et al. (2008a). Here, we summarise the key steps in developing a progressive terminology scheme that focuses on descriptive aspects first and then progresses to genetic terminology where possible. We also provide a ready reference flow chart or pocket guide (Figs. 1 and 2) that represents a distillation of the principal steps and tables of Cas et al. (2008a) and can be used in the field initially and then further if and when petrographic data becomes available. A key initial question in describing and assessing a particular facies is: is it coherent (i.e. not fragmented) or is it fragmental? If it is clear from the outset that the facies is coherent, then Fig. 1 and the succession of steps in it should be used. If the facies is clearly fragmental then Fig. 2 and the steps in it should be used. However, sometimes it is not clear if a facies is coherent or fragmental, perhaps because of the overprinting effects of alteration, or perhaps the presence of abundant lithic clasts,

R.A.F. Cas et al. / Lithos 112S (2009) 183–190 185

Fig. 1. Sequence of steps recommended in developing a progression from initial descriptive to final genetic terminology for coherent kimberlite facies or rock units. Steps are discussed in the text.

186 R.A.F. Cas et al. / Lithos 112S (2009) 183–190

Fig. 2. Sequence of steps recommended in developing a progression from initial descriptive to final genetic terminology for fragmental kimberlite facies or rock units. Steps are discussed in the text.

R.A.F. Cas et al. / Lithos 112S (2009) 183–190

which may suggest a fragmental character, but other features such as a uniform groundmass-like texture suggest a coherent character. In ambiguous cases such as this, both figures should be used initially until one decides that the facies is either coherent or fragmental, perhaps after examining thin sections. The steps are more or less similar in both figures, so major errors cannot be made in the initial stages using either, because those steps largely involve assessments of descriptive properties of deposits. 4. Coherent kimberlite (Fig. 1) 4.1. Step 1: assess the components and textures preserved The principal components of all igneous rocks can be subdivided into mineral crystals, lithic fragments and juvenile clast types, and the textures can be variable from inequigranular (i.e. large crystals dispersed in a fine groundmass) to equigranular (Fig. 1, Step 1). Large mineral crystals or grains could be magmatic phenocrysts (i.e. crystallised from the erupting magma) or xenocrystic in origin. It is now apparent from research of many different magma systems that many large crystals that previously were assumed to be phenocrysts could be xenocrystic. The general term “macrocryst/macrocrystic” can be used for all large crystals until it can be determined whether or not they are phenocrysts or xenocrysts. In terms of the textural elements, apart from large crystals and clasts, the fine interstitial material is a groundmass if the rock is coherent or matrix and/or cement, if the rock is fragmental. The term “groundmass” should be confined to coherent rocks and “matrix” to fragmental rocks to avoid confusion. Groundmass may contain vesicles or vesicles in-filled with secondary minerals, which are then called amygdales. However, vesicles need not be present at all. Distinguishing groundmass from matrix in altered rocks may be difficult, and since all kimberlites are altered, this is always a problem. A rule of thumb is that coherent groundmass usually has a uniform distribution of crystals and crystal populations that have reasonably consistent size, whereas crystals and crystal fragments in a fragmental rock could have irregular distributions and varying sizes. However, after documenting the various petrological components and textural features summarised in the flow chart for Step 1 in Figs. 1 and 2, it may still not be possible decide if the rock is coherent or fragmental (see Hayman et al., 2008, for a comprehensive discussion of the problem of distinguishing coherent from fragmental kimberlite). 4.2. Step 2: evaluate the type(s) and effects of alteration After the initial assessment of components and texture in Step 1 it is therefore essential to examine more closely the alteration mineral assemblages and the effects of alteration overprinting on both components and original emplacement texture (Fig. 1, Step 2). All kimberlites are altered to varying degrees (e.g. Stripp et al., 2006; Cas et al., 2008b; Hayman et al., 2008) to minerals such as serpentine, chlorite, carbonate, saponite, microlitic clino-pyroxene, etc. Most original minerals, including even opaques such as spinels, will exhibit some degree of alteration to secondary minerals. Such alteration is inevitable given that kimberlite magmas are ultrabasic and therefore highly chemically reactive. Natural fluids such as magmatic gases and hydrothermal fluids that permeate through a kimberlite during and immediately after an eruption, and even meteoric waters percolating through the kimberlite conduit and vent long after eruptive activity has ceased have the potential to react with the minerals and variably glassy groundmass and matrix components of kimberlites. However, the most important effect of alteration is the potential to overprint original textures, making it difficult to determine whether or not a rock was coherent or fragmental in the first instance, and thereafter the nature of the processes that emplaced the rock (see Cas et al., 2008b for a detailed discussion of these issues). Assessing the nature, intensity and effects of alteration are therefore an important

187

second step in documenting the deposit characteristics of kimberlite deposits, whether they are coherent or fragmental. In coherent rocks the principal effect of alteration is to overprint the original groundmass texture, including the nature of the groundmass (glassy or crystallised; Hayman et al., 2008). Serpentine alteration in particular is pervasive in its effects on the groundmass of coherent rocks. However, at times, there is a uniform distribution of small euhedral to subhedral olivine crystals preserved, which suggests that they were small phenocrysts in a coherent groundmass. The principal alteration minerals should be identified, and effects on original textures should be noted. Usually, this can only be done cursorily at the drill core or hand specimen scale, and requires detailed study of thin sections to fully understand how alteration has affected original textures (Stripp et al., 2006; Cas et al., 2008b; Hayman et al., 2008). 4.3. Step 3: apply the correct crystal size terminology There is no consensus in igneous petrology about the appropriate terminology for different crystal sizes. In sedimentology, the long established geometric progression of the universally accepted Wentworth grainsize scale is the basis for the grainsize scale, terms and size classes. For consistency, and to reduce the number of different size schemes to be remembered, we adopt a modification of the Wentworth grainsize scheme and its terms to describe different size classes of crystals (Fig. 1, Step 3). 4.4. Step 4: describe the crystal shape(s) Recognising crystal shapes may provide information about the origin of the crystals (xenocrystic versus phenocrystic), and the history of the crystals (quenched origin, resorption effects, fragmentation of crystals in the source or in transit, etc). Some common crystal form or shape terms are listed, but other commonly used terms can be applied where relevant. This is a strength of this terminology approach — it provides guidelines about principles, and gives examples (Fig. 1, Step 4), but other terms that are consistent with the guidelines can be applied where appropriate. 4.5. Step 5: assess the abundance of crystals Assessing the abundance of crystals in a coherent igneous rock can be done by visual estimation, by using visual comparator diagrams of abundances of components in a field of view, available in some petrography books. Alternatively it can be based on statistically accurate point counting, using grid counting in core or outcrop, a point counting petrographic stage for thin section analysis, by hand-picking or by mechanical (magnetic, heavy liquid) mineral separation techniques of crushed aggregates. Such analysis usually focuses on the macrocrystic crystal population, and the actual approach used will depend on the accuracy required by the study in hand. When particular mineral abundances, perhaps of important indicator minerals, are needed to a high level of accuracy the statistically most accurate methods are used. Fig. 1, Step 5, provides some general crystal abundance class names and their ranges. 4.6. Step 6: assess the vesicle abundance or vesicularity level of the groundmass Determining the vesicle abundance in a coherent rock or in pyroclasts can tell much about the volatile and exsolution history of a magma (e.g. Cashman, 2004). We have adopted the vesicularity class terminology and abundance range of Houghton and Wilson (1989). However, with low viscosity magma such as kimberlite, the final preserved vesicle content of the final lithified products are not necessarily indicative of the original volatile content of the magma, because the gas bubbles of exsolved volatiles can quickly escape from

188

R.A.F. Cas et al. / Lithos 112S (2009) 183–190

the magma during its rise to the Earth's surface in the conduit and even from pyroclasts after explosive fragmentation of the magma. 4.7. Step 7: compile a composite component, textural and compositional petrological name This step involves combining the key alteration type, components, textures and composition, into a composite descriptive petrological name. Guidelines about the order of assembling the terms to be combined are listed in Fig. 1, Step 7, together with some examples of appropriate names. 4.8. Step 8: note the unit thickness This step is applied irrespective of whether or not the coherent unit is a lava or intrusion. Again, a standard scale of thickness classes, with thickness ranges for each class, is listed (Fig. 1, Step 8). Although a standard classification of bedding thicknesses exists for sedimentary strata, that scheme is not applicable to coherent bodies because igneous bodies can be orders of magnitude thicker than sedimentary strata.

the slopes of an edifice or beyond the edifice (Fig. 1, Step 11). Some coherent kimberlites are of course blind intrusions, totally unrelated to a volcanic vent and edifice system. 4.12. Step 12: compile a final descriptive and genetic facies/deposit/rock unit name In those studies where a complete volcanological reconstruction is the aim, this represents the final step in applying a comprehensive, combined descriptive and genetic name to the various facies or rock units involved. In some studies this may not be required and just a practical descriptive rock name as discussed in Step 7 above may suffice. Or, just a totally interpretive or genetic name may be desired at this stage. In our experience, combining some of the key descriptive characteristics of a particular facies with its genetic origin name and setting helps to clearly characterise particular facies and distinguish them from other facies that may have some common components or textural features, but also other distinctive ones. Again, guidelines on the order of assembling the composite name are suggested and examples of possible facies names are listed in Fig. 1, Step 12. 5. Fragmental kimberlite (Fig. 2)

4.9. Step 9: record emplacement features/structures In this step important characteristics of the facies or rock unit that may help to understand the emplacement process are recorded. In coherent units this could include primary features such as flow banding, columnar jointing, autoclastic brecciation (see Cas and Wright, 1987, for discussion of characteristics), marginal apophyses into adjacent units, or post-emplacement features such as faults (Fig. 1, Step 9). 4.10. Step 10: assess contact relationships This step requires careful examination of the contact relationships between inter-facing units. First the contact should be assessed descriptively in terms of first order geometrical relationships (concordant or discordant with adjacent units and any internal primary layering or bedding they may display). The aim is to try to determine the general emplacement category (intrusive versus extrusive), and whether or not the contact is consistent with a conformable, supracrustal contact, an intrusive contact or a faulted contact (Fig.1, Step 10). These origins are not simple to determine in old rock successions, especially where the three-dimensional architecture and relationships must be reconstructed from drill core, and where alteration has masked the original nature of contacts. This can be very difficult. 4.11. Step 11: note the setting The reason for assessing the setting late in the process of developing terminology, has been outlined above, in Section 2, point 4. Too many kimberlite classification and terminology schemes relate facies to an expected position in the standard southern African pipe model (root, diatreme, crater), with expectations of what should be found. As noted, not all pipes are “model” pipes, and crater zone facies could be found in the diatreme, and root zone “hypabyssal” or coherent intrusions could be found in the diatreme or the crater zone. Furthermore, the prairies or Fort a la Corne bodies are not necessarily “pipes” but the preserved remains of positive relief edifices such as tuff cones. Assessing the setting therefore should only be done after the whole facies and stratigraphic architecture and palaeoenvironmental setting of the system has been reconstructed, and the emplacement processes have been evaluated. This step should involve an assessment of whether or not the setting of the body was subaerial or subaqueous, and whether or not the facies in question was emplaced within the vent (intra-vent), and at what level, or outside the vent (extra-vent), on

As for coherent kimberlite facies, we have developed a sequence of steps that progress from descriptive to genetic in developing an approach to terminology for fragmental kimberlite. Many of the steps are similar, but of course there are important differences between coherent and fragmental facies in terms of both descriptive characteristics and origins and so some steps are different (Fig. 2). 5.1. Step 1: document components and textural elements In fragmental kimberlite the principal components will be angular to rounded mineral crystal fragments, lithic fragments, and if the facies is pyroclastic, juvenile or magmatic pyroclasts (Fig. 2, Step 1). Even mass flow resedimented or tractionally reworked facies may contain juvenile pyroclasts that are transported by surface sedimentary processes, and are recognisable because they are fluidal or “spindle bomb” shaped, or highly vesicular in nature like pumice or scoria, or spherical pelletal lapilli. Accretionary lapilli are distinctive spherical aggregates of fine ash that commonly indicate the influence of phreatomagmatic explosive activity. The textural elements of fragmental kimberlite, just like any sediment can include large grains or clasts, called the framework grains or clasts, and fine particles in between, called the matrix. In deposits with an initial porosity, a chemically precipitated cement that has filled pore space may also be present. However, in altered fragmental rocks, determining if a cement or fine matrix was originally present between framework grains can be very difficult. Cas et al. (2008a) have indicated that in an altered fragmental rock with suspected original fine matrix, a good proxy for the fine matrix is a framework population of grains that is in itself poorly sorted, with a wide ranging grain size range down to that expected of fine matrix grains. 5.2. Step 2: determine the type of alteration, and the effects of alteration on original components and textures Determining the degree of alteration in fragmental kimberlite is very important so that the nature of the original depositional texture, which reflects the nature of the physical processes, can be determined. This is not only important academically but has applied significance in terms of understanding physical sorting processes that may influence diamond grades. It is therefore important to study and document the types of alteration minerals and their effects on original components and textures, after the initial assessment of components in Step 1. In fragmental kimberlites original clay and silt size grains in the fine

R.A.F. Cas et al. / Lithos 112S (2009) 183–190

matrix are almost never visible because they are almost invariably overprinted and replaced by serpentine. It is therefore not possible to easily assess how much original fine matrix there was, which is important for understanding the sorting characteristics and therefore the nature, intensity and sorting efficiency of the fragmentation, transportation and depositional processes and agents. If the framework grains (the large grains in a fragmental deposit) are closely packed to the point of being framework or clast supported, and about the same size, then it is unlikely there was much fine matrix deposited, indicating efficient sorting processes. If the deposit or facies is diamond bearing then it implies larger diamonds may be found whereas microdiamonds may be less common. In this case the sparsity of microdiamonds is not necessarily a good indicator of diamond grade, especially of macro-stones. However, if the framework grains are themselves highly variable in size and often separated by domains of fine interstitial material (a matrix supported texture) such as serpentine then the interstitial material must have been a fine clastic matrix to support the “floating” large grains in that matrix before the matrix was altered to serpentine. Such a facies may contain a greater range of diamond sizes, including microdiamonds. Being able to understand the nature of the alteration will enable understanding of original depositional textures which reflect on effectiveness of physical processes to effectively sort clastic grain populations during eruption and deposition. This is not easy and requires substantial petrographic work and understanding of clastic textures, physical processes, diagenetic and alteration processes and textures. 5.3. Step 3: apply descriptive grainsize terminology Cas et al. (2008a) have pointed out the difficulty in determining the appropriate grainsize terminology that should be applied in the early descriptive stages, before the origins of the facies have been determined. Applying grainsize terms for pyroclastic deposits at this stage would be misleading if the deposit turns out to be a resedimented volcaniclastic sediment. Similarly using familiar sedimentary size terms could also be misleading. However, because the Wentworth grainsize scale is applied universally for clastic aggregates in a number of disciplines, and without wanting to invent a new size terminology, Cas et al. (2008a) suggested using the sedimentary terms as an initial working terminology, and adding the suffix “-size” to each size term (e.g. sand-size, or boulder size breccia), to imply a size affinity without necessarily implying a specific genetic origin (Fig. 2, Step 3). A more genetic terminology can then be applied later in Step 11 if the genetic origins of the facies have been determined. 5.4. Step 4: assess the sorting level and textural class There are standard sorting class terms used to describe the sorting characteristics of sediments, and useful visual comparator diagrams of the characteristics of these sorting classes are commonly used (Fig. 2, Step 4). As outlined above, assessing sorting levels is important in terms of understanding physical processes and diamond grade patterns, but can be difficult due to the overprinting effects of alteration, which can mask original textural characteristics (see discussion to steps 1 and 2). Dunham (1962) proposed useful textural terms to describe different textural assemblages in clastic carbonates transported, variably physically sorted and deposited by currents in marine environments. The terms can therefore be just as usefully applied to other clastic aggregate types because the textures they represent are the outcomes of physical sorting processes that operate anywhere loose debris is being transported. The composition of the debris is incidental. In the Dunham scheme, a mudstone consists of N90% clay size fragments, reflecting low energy conditions. A wackestone is a matrix supported aggregate with dispersed larger fragments in a clay size matrix. A short-lived mudflow, debris flow or pyroclastic flow event would produce a wackestone texture.

189

A packstone is grain or framework supported with b5% clay size matrix, indicating reasonably energetic conditions, as occurs in a river or a turbidity current. A grainstone is a well sorted, grain or framework supported deposit without clay matrix, indicating very high-energy and efficient sorting conditions such as on a beach or during a pyroclastic fallout event. 5.5. Step 5: assess the degree of rounding The assessment of rounding should use standard sedimentary rounding–angularity terms (well-rounded, sub-angular, etc). Rounding usually reflects sustained abrasion by high-energy transport agents. However, in kimberlites rounded olivine crystals are usually the product of magmatic resorption in the magma conduit prior to eruption, not physical abrasion. 5.6. Step 6: estimate the free crystal content of the facies The same abundance classes and ranges are used as for coherent kimberlite (Fig. 2, Step 6). 5.7. Step 7: assess the vesicularity of juvenile clasts The same vesicularity classes and vesicle abundance ranges are used as for coherent kimberlite, using the scheme proposed by Houghton and Wilson (1989) for pyroclasts (Fig. 2, Step 7). As noted for coherent kimberlite however, the preserved level of vesicularity of kimberlite pyroclasts may not have the same volcanological significance as for more viscous magmas because of the ease with which gas bubbles can escape from low viscosity magma such as kimberlite and fragmented pyroclasts of kimberlite magma. Nonetheless to understand such processes it is important to document the characteristics. 5.8. Step 8: compile a composite descriptive component, textural, compositional, petrological name Again this step involves combining the major descriptive characteristics of the fragmental facies in question in a structured sequence (Fig. 2, Step 8). Examples of possible descriptive petrological names are also provided in Fig. 2, Step 8. 5.9. Step 9: assess bedding or depositional unit thickness Understanding bedding thickness helps to understand the magnitude of depositional events, the physical properties of the depositional agents, and the constraints of depositional topography. We adopt a long-standing scheme used in stratigraphy and sedimentology to name strata of different thickness (Fig. 2, Step 9). 5.10. Step 10: document depositional structures Using the principles of sedimentology and physical volcanology, depositional or sedimentary structures reflect most clearly the nature of the transporting and depositional agent and the conditions under which deposition occurred. Processes can be particulate, tractional or involve mass flow or movement (Fig. 2, Step 10; see Cas and Wright, 1987 for a comprehensive discussion of physical transport processes and deposit characteristics). Combined with the preserved original texture, this step will most closely allow interpretation of the causal physical transport process and the conditions. 5.11. Step 11: apply appropriate genetic size terminology Having determined, if possible, the emplacement origins of the facies in question, the appropriate genetic terms can now be applied to replace the descriptive terms applied earlier. Well-established size

190

R.A.F. Cas et al. / Lithos 112S (2009) 183–190

term schemes exist for pyroclastic and epiclastic sedimentary successions, but less formalized schemes exist for autoclastic deposits (Fig. 2, Step 11; Cas et al., 2008a). 5.12. Step 12: determine contact relationships The aim is to establish if contacts are conformable, unconformable, intrusive or faulted (Fig. 2, Step 12), to fully understand timing relationships. Contacts could be original or modified by faulting, which in the settings involved could be syn-depositional, and/or related to compactional and gravitational processes. 5.13. Step 13: determine setting Again, this involves understanding if the setting was subaerial or subaqueous, and intra-vent or extra-vent, based on a thorough palaeoenvironmental analysis and facies architecture reconstruction. 5.14. Step 14: compile a final genetic facies/unit/deposit composite name As for the coherent facies, guidelines and examples are provided in Fig. 2, Step 14. 6. Conclusions The proposed terminology scheme and “pocket guide” represent the most comprehensive and structured scheme yet proposed in kimberlite geology (and volcanology in general). It allows terminology to be developed progressively, from initially descriptive to finally genetic, using clearly explained guidelines. The scheme can be used in its entirety in both industry and academia, or just the more descriptive parts can be adopted, especially in industry, where many of the descriptive characteristics discussed here are already documented in day-to-day logging. Another practical outcome of the scheme is that is serves as a checklist of key observations that should be made and recorded, and that will then serve as the basis for process interpretations of the origins of each facies that is documented. Most importantly we advocate abandoning some terms that are traditionally used in kimberlite geology that don't have a specific, literal descriptive meaning, are vague, have developed incorrect connotations, and are not used in mainstream volcanology. In particular, the terms hypabyssal, autolith, tuffisitic kimberlite and tuffisitic kimberlite breccia, magmatic, magmaclast, and magmaclastic, should be abandoned. Although this may be difficult for some people who have used these terms regularly for a long time, we recommend their abandonment in the interests of reducing ambiguity and uncertainty in terminology and understanding of processes in kimberlite geology. Abandoning such terms and following the approach to terminology advocated here would bring kimberlite geology into much closer alignment with mainstream volcanology, which is long overdue given that kimberlite bodies are the remains of ancient volcanoes. Acknowledgements The Monash Kimberlite Research Group gratefully acknowledges the research funding support provided by BHP Billiton, De Beers Canada, Tahera Diamonds, Northwest Territories Canada Research, Kimberley Diamonds, Argyle Diamonds and Monash University research funding to undertake research into the volcanology of kimberlite and lamproite pipes. We thank reviewer Chris Smith and guest editor Steve Foley for their helpful suggestions for improving the manuscript.

References Cas, R.A.F., Porritt, L., Pittari, A., Hayman, P., 2008a. A new approach to kimberlite facies terminology using a revised general approach to the nomenclature of all volcanic rocks and deposits: descriptive to genetic. Journal of Volcanology and Geothermal Research 174, 226–240. Cas, R.A.F., Hayman, P., Pittari, A., Porritt, L., 2008b. Some major problems with existing models and terminology associated with kimberlite pipes from a volcanological perspective, and some suggestions. Journal of Volcanology and Geothermal Research 174, 209–225. Cas, R.A.F., Wright, J.V., 1987. Volcanic Successions. Allen and Unwin, London. Cashman, K.V., 2004. Volatile controls on magma ascent and degassing. The State of the Planet: Frontiers and Challenges in Geophysics vol. 150, vol. 150. American Geophysical Union Monograph, pp. 109–124. Clement, C.R., 1982. A comparative geological study of some major kimberlite pipes in the North Cape and Orange Free State. Unpublished PhD thesis, University of Cape Town. Clement, C.R., Skinner, E.M.W., 1985. A textural–genetic classification of kimberlites. Transactions Geological Society of South Africa 88, 403–409. Dunham, R.J., 1962. Classification of carbonate rocks according to depositional texture. In: Ham, W.E. (Ed.), Classification of Carbonate Rocks. American Association for Petroleum Geology Memoir vol. 1, vol. 1, pp. 108–121. Field, M., Scott Smith, B.H., 1998. Textural and genetic classification schemes of kimberlites: a new perspective. Extended Abstracts of the Seventh International Kimberlite Conference, Cape Town, South Africa, 1998, pp. 214–216. Field, M., Scott Smith, B.H., 1999. Contrasting geology and near surface emplacement of kimberlite pipes in southern Africa and Canada. In: Gurney, J.J., et al. (Ed.), Proceedings of the seventh International Kimberlite Conference, Cape Town. Red Roof Design, pp. 214–237. Hayman, P.C., Cas, R.A.F., Johnson, M., 2008. Difficulties in distinguishing coherent from fragmental kimberlite: a case study of the Muskox pipe (Northern Slave province, Nunavut, Canada). Journal of Volcanology and Geothermal Research 174, 139–151. Hayman, P.C., et al., 2009, this issue. Characteristics and alteration origins of matrix minerals in volcaniclastic kimberlite of the Muskox pipe (Nunavut, Canada). Proceedings of the 9th International Kimberlite Conference. Lithos 112S, 473–487. Houghton, B., Wilson, C.J.N., 1989. A vesicularity index for pyroclastic deposits. Bulletin of Volcanology 51, 451–462. McPhie, J., Doyle, M., Allen, R.L., 1993. Volcanic Textures. University of Tasmania. Mitchell, R.H., 1986. Kimberlites: Mineralogy, Geochemistry, and Petrology. Plenum Press. Mitchell, R.H., 1995. Kimberlites, Orangeites and Related Rocks. Plenum Press. Nowicki, T., Crawford, B., Dyck, D., Carlson, J., Mcelroy, R., Oshust, P., Helmstaedt, H., 2004. The geology of kimberlite pipes of the Ekati property, Northwest Territories, Canada. Lithos 76, 1–27. Pittari, A., Cas, R.A.F., Lefebvre, N., Robey, J., Kurszlaukis, S., Webb, K.J., 2008. Eruption processes and facies architecture of the Orion Central kimberlite volcanic complex, Fort à la Corne, Saskatchewan; kimberlite mass flow deposits in a sedimentary basin. Journal of Volcanology and Geothermal Research 174, 152–170. Scott Smith, B.H., Nowicki, T.E., Russell, J.K., Webb, K.J., Hetman, C.M., Harder, M., Mitchell, R.H., 2008. Kimberlites: descriptive geological nomenclature and classification. Extended abstracts. Ninth International Kimberlite Conference. No. 9IKC-A-00124. Skinner, E.M.W., Clement, C.R., 1979. Mineralogical classification of southern African kimberlites. In: Boyd, F.R., Meyer, H.O.A. (Eds.), Kimberlites, Diatremes and Diamonds: Their Geology, Petrology and Geochemistry. 2nd International Kimberlite Conference Proceedings. American Geophysical Union, Washington, pp. 129–139. Smith, C.B., 1983. Pb, Sr, Nd isotopic evidence for sources of southern African Cretaceous kimberlites. Nature 304, 51–54. Smith, C.B. and Lorenz, V., 1986. Volcanology of the Ellendale lamproite pipes, Western Australia. In Ross, J. (ed) Kimberlites and Related Rocks. Volume 1. Their Composition, Occurrence, Origin and Emplacement. Proceedings of the Fourth International Kimberlite Conference, Perth, 1986. Geological Society of Australia Special Publication 14, 505–519. Sparks, R.S.J., Baker, L., Brown, R.J., Field, M., Schumacher, J., Stripp, G., Walters, A., 2006. Dynamical constraints on kimberlite volcanism. Journal of Volcanology and Geothermal Research 155 (1–2), 18. Sparks, R.S.J., et al., 2009, this issue. The nature of erupting kimberlite melts. Proceedings of the 9th International Kimberlite Conference. Lithos 112S, 429–438. Stripp, G.R., Field, M., Schumacher, J.C., Sparks, R.S.J., Cressey, G., 2006. Postemplacement serpentinization and related hydrothermal metamorphism in a kimberlite from Venetia, South Africa. Journal of Metamorphic Geology 24, 515–534. Woolley, A.R., Bergman, S.C., Edgar, A.D., Le Bas, M.J., Mitchell, R.G., Rock, N.M.S., Scott Smith, B.H., 1996. Classification of lamprophyres, lamproites, kimberlites, and the kalsilitic, melilitic, and leucitic rocks. Canadian Mineralogist 34, 175–186. Zonneveld, J.-P., Kjarsgaard, B.A., Harvey, S.E., Heaman, L.M., McNeil, D.H., Marcia, K.Y., 2004. Sedimentologic and stratigraphic constraints on emplacement of the Star Kimberlite, east central Saskatchewan. Lithos 76, 115–138.

All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.