Abstract
One of the remarkable characteristics of the Neuquén Basin is that a group of Cenozoic shallow microgabbroic sills exhibit gas and oil shows, and some have produced hydrocarbons for up to 20 years. Their capacity to act as reservoirs is caused by the increase in the effective porosity owing to the simultaneous development of petrological and fracture processes during cooling that generate systematic variations inside the intrusive bodies.
In this work we describe in detail an unusual intracolumnar joint system formed by two joint sets, referred to here as concentric and radial joint sets. This atypical joint system is developed within the small stacked polyhedrons into which certain zones of classic prismatic columns are transformed during the cooling processes. These small polyhedrons are laterally bounded by the vertical planes that limit the columns, and at the top and bottom by transversal joint planes oriented almost perpendicular to the column axis.
As a result of detailed macroscopic analysis, the genesis of the concentric joint sets have been interpreted as resulting from either the development of natural convective flows or from complex processes of cooling influenced by the joint-bounded planes of the columns. When the radial joint set is present, it is clearly related to the development of thermal stresses between the outer and inner parts of the small polyhedrons.
In geology textbooks and previous published papers the genesis of what is informally called ‘onion skin’ joint sets in basic intrusive and extrusive igneous rocks has been attributed to processes related to either weathering or deuteric alteration. However, based on detailed macroscopic observations, our work suggests that, in most cases, any type of alteration process only accentuates the planes previously generated during the cooling process.
Geological observations carried out during the early decades of the last century by Groeber (1929) in the Neuquén Basin, Argentina, and Wade (1929) in Madagascar, Africa, suggested that, in those areas with abundant Cenozoic igneous activity, the heat transferred from intrusive rocks to organic-rich host sedimentary rocks triggered the generation of hydrocarbons. Today, it is widely accepted that intrusive igneous rocks can play a variety of roles in the processes of generation, migration and trapping of hydrocarbons.
Related to the hydrocarbon-generation processes, igneous intrusions represent one of the possible heat sources that thermally alter the organic compounds present in the sedimentary rocks. A temperature peak over a relatively brief time following a fast maximum-temperature reaction can create thermal maturity (Robert 1980; Barker 1983; Kisch 1987).
Later, Barker (1988) pointed out that the heat flow through deep fractures and igneous intrusions throughout the Tertiary caused petroleum generation to occur in the western area of the Delaware Basin, USA. According to Johnson (1989), the heat pulses caused by the Tertiary intrusive igneous rocks in the Piceance Basin, Colorado, USA, increased the speed of transformation of the petroleum into gaseous hydrocarbons. In the Triassic Hartford and Newark basins, USA, Pratt et al. (1988) related the sharp variation in the degree of maturity of the black shales, which act as source rock in a stratigraphic interval of less than 1 km, to a high heat flow caused by basic igneous intrusions during the early Jurassic.
Most studies concerning hydrothermal systems agree that the rise in temperature in the area surrounding the igneous intrusion is what enables hydrocarbon generation to occur in only a few thousand years (Barker 1988; Didyk & Simoneit 1989). However, hydrothermal fluids link the generation with migration processes, transporting hydrocarbons at least in the first stages of the migration. In the same way, Hulen et al. (1994) described a geothermal system that was able to create an economically significant oil field.
Even though there are many references of bitumen seeps and oil and gas shows related to basic sills in many basins all around the world (Schutter 2003), references about basic shallow intrusive igneous rocks acting as reservoirs of hydrocarbons with commercial productions are notably scarce. One of the best-known examples is the fractured syenitic sill of the Dineh-bi Keyah field in Arizona (McKenny & Masters 1968). In this case, as with other well-known examples, the reservoirs have been assigned to the development of secondary fracture porosity. Recently, Wu et al. (2006) has included the presence of steeply dipping cracks and dilatational fractures of primary type present in a dolerite body of Jiyang Basin, China, as another necessary quality to constitute a reservoir.
The purpose of this paper is to describe and interpret the genesis of systematic variations in the fracturation system within microgabbroic sills, and to show that these fracturation systems are strongly related to cooling processes. In addition, intracolumnar, concentric and radial joint systems are described in detail. In some zones within the sills these atypical joint systems allow primary porosities of around 15% and a range of permeabilities of between 0.1 and 100 mD, enabling them to act as reservoirs, some of which have commercial viability.
Geology and tectonic setting
The Neuquén Basin is one of the largest oil- and gas-producing basins in Argentina. It is located in the mid-western part of the country (38°S, 69°W), and covers an area of approximately 160 000 km2 (Fig. 1).
Map of the Neuquén Basin showing the main tectonic areas, and the location of gas and oil producing fields. Filomena, Cura Pachingo, Aguada San Roque and Loma Las Yeguas shallow basic sills location are also shown. Line a–b indicates the direction of the cross-section for Figure 2.
The sedimentary sequence was affected by several tectonic events during Mesozoic and Tertiary times, and as a result two different morphological–structural areas are distinguished: the Andean Area, a fold and thrust belt in the western part of the basin that contains the active Andean Volcanic Arc; and the Extraandean Area, in the eastern and the central parts of the basin and platform areas, where fault-controlled block basement tectonics predominate (Fig. 1).
In Upper Triassic times, a group of extensional troughs were formed and then filled with volcanic and continental clastic deposits. Since then the Neuquén Basin has been a region of cyclical marine – continental sedimentary accumulation. From the Lower Jurassic to the Lower Tertiary, several marine – continental cycles developed with their depositional systems controlled by relative sea-level changes and/or tectonic events within the basin (Vergani et al. 1995).
From the Lower Jurassic to the Lower-Late Cretaceous, three main petroleum systems developed (Urien & Zambrano 1994). Their names were derived from the formation name of the source rock, followed by the name of the major reservoir rock. They are, in chronological order: Los Molles–Lajas; Vaca Muerta–Tordillo; and Agrio–Avilé (Fig. 2).
Geological section west–east through the Neuquén Basin illustrating the three main petroleum systems and the igneous activity related to the emplacement of gabbroic sills and granodioritic laccoliths, which act as reservoirs. Stars and arrows mark the main migration paths.
According to Villar et al. (2005), source rocks from the Los Molles, Vaca Muerta and Agrio formations commonly include basinal and outer-shelf marine-rich organic black shales, each with different potentials for hydrocarbon generation.
Clastic rocks of the Lajas, Tordillo and Avilé formations are mainly sandstones deposited in shallow-water deltaic and marine environments (Maretto et al. 2002; Masarik 2002; McLlroy et al. 2005). These sandstone bodies, together with shallow-marine carbonate rocks, constitute most of the reservoir rock. Thick bodies of evaporites, mainly gypsum and halite, frequently act as regional seals.
Cenozoic igneous activity within the Neuquén Basin
For the past 65 Ma, the evolution of the active Andean Convergent Margin has influenced both the timing and the distribution of igneous activity. Subduction of the oceanic Nazca Plate underneath the continental South American Plate has generated a magmatic arc that extends roughly parallel to the margin (Fig. 3).
Schematic cross-section across the active convergent Andean Margin at 38°S latitude where the Neuquén Basin is located. This sketch shows the different igneous sources of gabbroic sills intruded in the back-arc area.
At the surface, the associated volcanism defines a volcanic arc, the width of which has changed several times in an east–west direction. These processes have been interpreted by Kay et al. (2006) as a consequence of changes in the age and angle of the subducting oceanic plate.
In the subsurface, the expression of this associate volcanism is reflected in a series of shallow intrusive bodies, with variable composition and geometric shapes that form laccoliths of dioritic and granodioritic compositions, gabbroic sills, as well as acid and basic dyke swarm complexes (Fig. 2).
The group of basic sills studied in this work intrudes black shales from the Upper Jurassic – Lower Cretaceous Vaca Muerta Formation and have been K/Ar dated with absolute ages of 23.6 ± 4.4 Ma (Aguada San Roque sill), 12.5±1.0 Ma (Loma Las Yeguas sill), 12.5±1.1 Ma (Filomena sill) and 7.3±2.2 Ma (Cura Pachingo sill).
On the basis of thin-section studies, all sills have been classified as microgabbros. Grain size is typically less than 1 mm, with the average grain size increasing up to 2 mm towards the interior of the sill, especially in sectors where differentiates are present. The texture is subhedral – granular, but occasionally subophitic texture predominates. Microgabbros are composed of up to about 50% of normal zoned crystals of labradoritic plagioclase (An58/63).
The relative proportion of dark- to light-coloured minerals is nearly 40%. The main ferromagnesian mineral is a pale green–pinkish brown clinopyroxene, identified petrographically as titaniferous augite. Typically, subhedral and anhedral augite partially enclose plagioclase crystals, forming the classical subophitic texture (Fig. 4a). Small intercrystalline spaces between tabular plagioclase crystals are occupied by analcime and/or albite. Accessory minerals identified as magnetite and apatite also form tiny needles within the plagioclase crystals.
(a) Photomicrographs showing the subhedral–granular to subophitic texture, which is fairly typical in microgabbroic sills. Px, piroxene; Pl, plagioclase. Crystal size is 1 mm. (b) Small differentiate syenitic lenses from a drill core of the Cura Pachingo sill.
Leucocratic differentiates classified as syenites are lighter coloured than microgabbros owing to the mafic mineral content decreasing to between 25 and 15%. Accordingly, plagioclases are calcic andesines (An40/50). Syenitic differentiate bodies occur as small masses and form: magmatic layering, ritmic graded layers, and network zones with veins and veinlets (Fig. 4B).
Both microgabbros and syenites display deuteric alteration as overgrowths by chlorites and biotites on piroxene crystals. Microgabbro SiO2 content ranges from 45 to 51%, and total alkalis (Na2O+ K2O) from 5.7 to 2.52% (Table 1). In the classification diagram proposed by Wilson (1989), all samples of the sills plot within the alkaline gabbro field, except for the Loma Las Yeguas sill differentiate which plots in the syenite field (Fig. 5).
Chemical classification and nomenclature of plutonic rocks from Wilson (1989). Curved line subdividing rocks into alkaline and subalkaline series is from Irvine & Baragar (1971). Cura Pachingo sill: CuP (▴) (average of five samples); Loma Las Yeguas sill: LLY (•); Aguada San Roque sill: ASR (◯); and Filomena sill: Fi (△) (average of six samples).
Representative major (wt%) element abundances in the basic intrusive rocks
Isotopic values of 87Sr/86Sr are compatible with the classic sources and petrological processes developed in this segment of the active Andean Convergent Margin. Those that have values of 87Sr/86Sr=0.7064 are in agreement with the low degrees of partial melting from a garnet-rich mantle source, while others with values of 87Sr/86Sr varying from 0.70348 to 0.70371 could represent a variable subduction component (Fig. 3).
Gabbroic basic sills acting as reservoirs
The fortuitous appearance in 1978 of gas reservoirs in intrusive basic igneous rocks within the Neuquén Basin immediately motivated the identification of these types of rocks by means of seismic reflection profiles. Orchuela et al. (1989) identified these igneous bodies as sills owing to their flat display, and concordant and gently inclined high-amplitude reflectors standing out clearly against a background of more diffuse reflectors of sedimentary succession. In cross-section, typical seismic expressions of sills are represented as flat bodies, with length values that exceed their thicknesses significantly.
In plan view, sills have an almost ellipsoidal shape, with maximum diameters ranging from 8 to 13 km. Sections taken normal to the maximum length show that feeder fractures are always located at the maximum relative depth and that the sills ascend from this point towards higher stratigraphic intervals. Changes or ‘jumps’ in the levels of intrusion are typically induced by lithological facies changes or previous faults (Fig. 6).
Seismic section showing a typical cross-section of a sill. Light grey colour represents the sill, and thin black lines previous and synintrusive faults.
Palaeodepths of the sills have been estimated to be 1500–2000 m, depending on the thickness of the overlying sedimentary column. The sills range in thickness from 18 to 45 m. These thicknesses were determined from drill-cutting descriptions and from their low gamma-ray log values (20–40 gAPI; gAPI=American Petroleum Institute gamma ray units) in contrast to the host black shales.
Magma densities at the time of the emplacement, calculated according to their geochemical composition (Botinga & Weil, 1972), range between 2.79 and 2.86 kg m−3.
Primary and secondary porosity
Effective porosity and permeability are essential conditions to develop reservoirs in basic shallow intrusive igneous rocks; i.e. interconnected pore spaces as well as permeability that allows the storage and subsequent circulation of fluids is required. Therefore, an understanding of the types of porosity of basic sills is critical for their evaluation as potential reservoirs. The widely accepted typical classification of the pore space for sedimentary rocks includes two kinds: primary porosity – which is originated in the initial processes that deposited the sediments; and secondary porosity – which result from processes that occur after the rocks have been formed. Likewise, this classification is used in this paper to define the type of porosity in basic shallow intrusive igneous rocks; however, the definition is adjusted according to the specific characteristics of these rocks. The proposed classification for the different types of porosity and genesis for basic shallow intrusive rocks is shown in Table 2.
Types of porosity in mafic shallow intrusive rocks
It is assumed in the case of shallow intrusive basic igneous rocks that the generation of primary pore spaces is a continuous process that begins with the rock emplacement and continues during the entire cooling period. Petford (2003) defines the primary porosity in a shallow intrusive basic igneous rock as the pore space originating during the emplacement and cooling processes. We add to his definition: ‘with no changes in its original whole mineralogical composition’. In this way we have labelled the different types of voids generated in the later stages of cooling by deuteric alteration as primary porosity. Also, we have divided primary and secondary porosity in two types: fracture and cavity porosity.
The primary fracture porosity is represented as columnar and intracolumnar jointed structures and flow jointing. Columnar joints are three-dimensional fracture networks that organize the solid rock in pentagonal and hexagonal columns that are formed, in this case, during the cooling process of shallow intrusive basic rocks. Joint tensional-fracture networks include joint-bounded columns and transverse-joint planes oriented normal to the column axes; the latter forming bands called ‘striae’ or ‘chisel marks’.
Intracolumnar primary porosity related to the cooling process is a concentric and radial joint system that develops within the columns. This joint system will be described in detail in this paper. Flow jointing is a type of primary fracture porosity related to magma viscosity and occurs when shears between the planes are generated. These types of planes are separated by a few millimetres and have been observed close to the contacts with the host rocks. Sometimes these planes are marked by microcrystals oriented parallel to the contacts or microvesicles.
Primary cavity porosity includes vesicles of millimetres–centimetres in size, generated by expanding gases that may be locally abundant, especially near the top of the sills. Primary cavity porosity includes intercrystal voids and miaroles that vary in size from centimetres to millimetres, and are generated during deuteric alteration processes that are widespread in zones with differentiated late-magmatic veins.
Secondary porosity is comprised of the pores that are developed in the last stages of the cooling process – and even after this process has finished – which motivate changes in the original mineralogical composition of the rocks. Such changes are usually caused by the alteration processes – which are triggered, in turn, by contact with hydrothermal fluids. Secondary porosity also includes the pores developed by tectonic activity that begins after the rock has been consolidated.
Summarizing, secondary porosity is mainly related to tectonic events and hydrothermal alteration processes. Tectonic events produce systematic joint systems that we call ‘tectonic joints’ and ‘fault and associated joint systems’ which affect the wall rocks as well as the sills that can play an important role in the migration of fluids to the sills.
Secondary porosity related to alteration processes is developed once the cooling process has finished and the rock comes in contact with hydrothermal fluids. The most common effects of hydrothermal alteration are: dissolution; replacement or elimination by partially or fully dissolving the original minerals, leaving behind voids or ‘boxworks’ with the precipitation of new minerals along fluid pathways; and microbrecciation.
In this way, intracrystalline and intercrystalline spaces are created. These voids are usually interconnected by microfractures, usually hydraulic microfractures or hydrofracturing, and as a consequence may give a considerable degree of permeability to basic igneous rocks.
So far, the best reservoirs in productive basic sills within the Neuquén Basin have effective total porosity (primary plus secondary porosity) averages that range from 10% to 15%, an average fracture permeability of 250 mD, and an average matrix permeability of 0.01–0.1 mD. Permeability may be due to either the primary or secondary pore network, or both (Perea et al. 1984).
The term matrix porosity is used to characterize all types of voids located inside the blocks that are limited by the biggest fractures with the maximum relative aperture. In the case of shallow basic intrusive rocks, matrix porosity may include cavities and microfractures (Table 2).
In addition, another form of porosity classification that may also be applicable to basic shallow intrusive rocks was established by Petford (2003). He proposed the classification of porosity in igneous rocks taking into account the capacity of fluid transport in ‘diffusive’ (Class D) and ‘flow’ porosity (Class F). Diffusive porosity includes the possibility of interfracture diffusion development. Consequently, intracolumnar concentric and radial joints sets, microfractures, as well as microfaults present in mafic sills would be included in this type of porosity. In addition, diffusive porosity could be thought of in a similar way as matrix porosity. Equally important, flow porosity includes all macrofractures that permit flow along the planes. In basic intrusive igneous rocks this may include joints limiting planes of columnar jointing, systematic tectonic joints and faults. As a consequence, flow porosity includes nearly all fracture porosities (Table 2).
Summing up, in basic shallow intrusive igneous rocks total porosity, i.e. the sum of primary and secondary porosity, is the result of mega-spaces (voids, vesicles, tectonic and cooling joints) and meso- and micro-spaces (vesicles, microfractures, and inter- and intracrystaline voids). Both primary and secondary porosity can include fracture and cavity porosity.
Internal development of the shallow intrusive microgabbroic sills
A transverse profile of a typical sill has been reconstructed using field analogues, borehole images and drilled core sections. Texture, petrological and geochemistry variations that develop within basic sill intrusions have been described in detail in geological literature for some time (McDougall 1962; MacLeod 1981; Froelich & Gottfried 1988). They have not, however, been compared with systematic variations in the fracturation system.
For the sills described in this work, systematic variations as mentioned in the above paragraph are in agreement with variations in the fracturation system.
Sills that intrude during one single magmatic phase generally consist of four zones arranged in a roughly symmetrical way from top to bottom, and they are identified by their particular texture, petrological structure and fracturation (Fig. 7). The relative thickness of each zone and its location can vary.
Illustration of the typical changes in texture and fracture density of a sill.
Contacts at the bottom and top of the sill against black shales are of planar sharp type and define Zone 1, which has an average thickness of a few centimetres. This zone represents a basaltic chilled margin with microporfiritic texture. A small amount of both plagioclase and pyroxene microphenocrysts are dispersed in the cryptocrystalline groundmass.
This thin zone frequently shows flow jointing determined by oriented microcrystals of plagioclase, and in some cases by the stretching of microvesicles. Flow-jointing planes are approximately parallel to each other and to the contact plane. They are thin with separations of 1 mm, but towards the centre of the sill their spacing gradually increases until they disappear. Viewed in cross-section each plane shows continuity up to several metres.
Stepping towards the centre of the sill, Zone 2, with a thickness that ranges from tens of centimetres to 2 m, is formed by fine-grained microgabbro with a microcrystalline texture and a development of columnar jointing with transverse joint planes separated by 30–50 cm. Further into the sill, Zone 3 consists of several metres of fine- to medium-grained granular gabbro with subophitic texture. The development of columnar joints continues, but with less separation (about 10 cm) of the transverse joint planes. This is the zone with the biggest fracture primary porosity owing to the development of intracolumnar jointing, i.e. concentric and radial joint system.
At the centre of the sill, Zone 4 defines a zone with granular medium-grained texture and thicknesses ranging from 1.5 to 2 m. The proportion of dark minerals decreases in this zone owing to the gabbro being more silicic, with abundant intracrystalline voids and miarolitic cavities (Fig. 8). This, and the evidence of magmatic diversification processes like layering, rhythmic graded layers and zones of net-veined syenites (59% SiO2) that develop in the final stages of crystallization, confirm the development of deuteric processes within the sill. This zone, usually located slightly above or below the central part of the sill, exhibits a lower density of transversal joints. Zone 4 can also develop with lesser thicknesses as intercalations in other parts of the sill. Small masses and transgressive bodies with limited lateral continuity are present in all zones of the sill.
Miarolitic cavity in a drill core of the microgabbroic Cura Pachingo sill (Zone 4). The irregular cavity is occupied by crystals of plagioclase crystallized in open spaces. These crystals are twice the size of the primary crystals of plagioclase in the microgabbro. In these zones primary porosity values can be up to 10–15%, but usually miaroles are discontinuous pore spaces.
Columnar jointing
The tiers that develop columnar joints appear to form a colonnade zone. The length of the columns and the diameters of the polygonal joint system are not easy to determine owing to the fact that the reconstruction has been made through limited observations in cores. Generally, as determined by indirect observations and field analogies, the column heights would vary between 3 and 5 m, and the diameters between 10 and 50 cm, depending on the total thickness of the sill.
One of the distinctive characteristics is the presence of very well marked joint planes oriented perpendicularly to the column axis. These joint planes are arranged in stacked succession of variable heights ranging between 5 and 30 cm, with the average around 10 cm (Fig. 9a). These transverse joint planes, in the case of the sills that have been studied, are straight or slightly curved, with both smooth and rough surfaces. These joint planes have been interpreted as successive cracks that form during discrete events of growth of the columns (Degrafff & Aydin 1987) accompanied by the cyclic stress of accumulation and release in the cooling body (Peck & Minakami 1968; Ryan & Sammis 1978).
(a) Borehole acoustic image from Zone 3 of the Cura Pachingo microgabbroic sill showing well-marked open transversal joint planes. Some of them display curved boundaries related to the development of intracolumnar concentric jointing structure. (b) Borehole resistive image from a microgabbroic sill of the Neuquén Basin exhibiting variations in electrical conductivity interpreted as concentric jointed structures.
Concentric and radial joint systems
In all the sills that are acting as petroleum and gas reservoirs, or have gas and oil shows, the joint systems described in the next paragraphs have a decisive influence on the development of higher values of effective porosity within a reservoir rock.
The development of zones with concentric joint sets controlled by vertical and horizontal joints was observed in borehole images (Fig. 9b), field outcrops (Fig. 10) and in core descriptions of the sills (Fig. 11).
Outcrops of Cenozoic microgabbroic sill few kilometres south of the Rio Grande oil field, Neuquén Basin. In cross-section, normal to its axis, these columns exhibits well-developed concentric joint sets around a massive core located in the centre of each column. Note the massive border, a few centimetres thick, that forms the external walls of the columns.
Concentric and radial joint systems in core of Loma Las Yeguas microgabbroic sill. The drill core sector to the left shows small stacked polyhedrons, each one with the development of a concentric joint set outside. To the right a sketch normal to the small polyhedron axis shows in detail the relationship between the concentric and radial joint sets.
In cores the small stacked polyhedrons that form the column have a predominant height of 10 cm and approximately the same diameter (Fig. 11a). Inside them, the development of a concentric joint system with separations between the concentric planes ranging from 1.5 to 3 mm has been observed.
Throughout the different sectors, the concentric planes are discontinuous and consequently winding, with several intersections between the planes of the different bands. The concentric joint planes are very thin and their surfaces are rough. They are cut by non-continuous planes with radial or pseudo-radial design that begins on the outer limits of the massive nucleus, located in the central zone.
The rounded massive nucleus is harder than the outer zone. It has a spherical or ellipsoidal shape and is clearly visible in the concentric zone because of its ‘egg-rock’ shape. Diameters of these ‘egg-rocks’ can vary between 2 and 4 cm. Many of them present processes of differentiation with respect to its exterior zone marked by the relative increment of light-coloured minerals. (Fig. 11b)
The numerous intersections between the open planes of the concentric and radial joint systems located inside the small polyhedrons cause the loss of cohesion in the rock, and, if a minimum stress is applied, the igneous rock behaves like ‘unlithified sand’. In analogous examples in the field, it has been observed that the separation between the concentric joint planes and the volume of the massive zone increases when the diameter of the polyhedrons and the separation distance between the transverse planes to the column also increases.
Interpretation of the processes
The observations described in the previous section suggest that the concentric joint systems that are developed inside the small polyhedrons begin their formation at the same time as the transverse planes. After two successive (top and bottom) transverse planes have been formed during the cooling process, one small polyhedron is generated. The height of the polyhedron, i.e. the distance between the transversal fractures, will be determined by the decreasing temperature gradient towards the inside of the igneous cooling body (Jagla & Rojo 2002). This polyhedral body is limited laterally by the edge planes that form the column (Fig. 12).
Interpretation of processes developed during the cooling in a jointed columnar structure. The convective processes of heat transfer that are developed between the edge planes of the columns cause temperature differences between the outer edge and the interior of the columns. This temperature gradient could induce heat transfer by conduction towards the outside of the columns and free convection processes within each one of the stacked small polyhedrons limited by transversal planes. These conductive and convective processes of heat transfer would have generated the concentric and radial joints system within the columns.
Once generated, this small polyhedron continues its process of cooling as an independent unit or ‘isolated cell’. Convective processes that reduce the temperature compared to the inside of the small polyhedron would be able to develop between the edges planes of the columns. Internally, the polyhedron is warmer, but loses heat by conduction towards the joint planes that limit it. The maximum temperature is located at the centre of the polyhedron, and the isothermal surfaces are concentric around this centre (Budkewitsch & Robin 1994).
The radial joints that cut across the concentric joint set within the polyhedrons are discontinuous planes oriented perpendicularly to the direction of maximum thermal tensile stress and are clearly related to the cooling process. This has been described by various authors, especially in fast cooling lavas such as volcanic bombs and pillow lavas (McDonald 1972; Yamagishi et al. 1989; Walker 1992). In this case the radial joints that cut across the concentric joint planes would represent the last phase of cooling. The close relationship between the concentric and radial joint systems inside each small polyhedron supports the idea that they were both generated during cooling processes.
In this paper two preliminary hypotheses are considered to explain the formation process of the concentric joints set:
Hypothesis 1
The concentric planes could be the expression of the free natural convective processes developed within each of the stacked segments that are forming the columns. These convective systems may have developed because of specific changes in densities and viscosities. Temperature variations between the external walls and the inside of the columns result in the former behaving like a solid, losing heat through conduction, whereas the latter behaves like a fluid.
The presence of differentiates in the central part of the concentric structure hold up the idea that convective processes with movements of small-scale fluids can be produced even inside the small polyhedron.
Spera et al. (1995) carried out systematic simulations using a mix of silicate melts with the purpose of studying heterogeneities in magmatic bodies. In simulations of bodies that are simultaneously cooled from the sides and above, and where the length of the body is much greater than the width, basic conditions tend to develop on the outer margins and more acidic conditions develop towards the centre, i.e. simultaneous convection and phase changes. These authors point out that the phenomena described by the model may occur in a wide-ranging scale of macroscopic lengths from millimetres to kilometres.
In addition, concentric flow patterns have been observed in laboratory experiments of free or natural convection where convective flows are developed within vertical wall-bounded slots, and the convection takes the shape of ‘multiple vertical cells’; in other words, cells located one on top of the other. Also, similar shapes have been obtained in convection cells called Rayleight–Bernard cells, which adjust to the containing walls in a rectangular box filled with silicone oil. This box is evenly heated from below and the convection in this case is due to buoyancy (Leeder & Pérez 2006, p. 200, fig. 4.155).
Transportation of fluids inside the material mass, instead of throughout the polygonal system of fractures, has been documented in laboratory experiments during the formation of columnar jointing (Goehring & Morris 2005). The concentric joint set could represent, in this way, the ‘marks’ of convective processes that have taken place inside the small polyhedrons.
Hypothesis 2
The curved concentric joints would approximately reproduce the design of the isothermal surfaces, and the concentric planes could be the result of remarkable variations in the thermal stress during the cooling process, limited by the polygonal shape of the columns (Fig. 12). Curvilinear joints with quasi-concentric design have been observed at microscopic scale in minerals, and their development attributed to thermal stress influence during the cooling of lava flows (Samanta 2001).
Conclusions
One of the interesting characteristics of the Neuquén Basin of Argentina is that some Cenozoic shallow basic sills of palaeodepths ranging from 1 to 3 km have been acting as hydrocarbon reservoirs for up to 20 years.
Sill-effective fracture primary porosity is linked to cooling processes that create the typical columnar jointing, and also, in some zones within the sills, an intracolumnar joint system. This system has few references within geological literature and to date has received little study. This joint system, referred to here as a concentric and radial joint system, increases effective porosity such that when drilled the sill exhibits reservoir properties similar to those of unlithified sand.
During the cooling process different zones are formed within the sill, and in some of these zones well-formed polygonal columns are bisected by transversal joints that are oriented normal to the column's axis. Then, during the same episode of cooling, concentric and radial joint sets are formed in a sequential manner within these discrete bisected polygonal columns. Concentric joint planes exhibit marcations owing to natural free convective flows that originated during the magma cooling processes, and are a result of relative differences in density and temperature within the columns.
The radial joint planes present in some of the studied sills originate from the centre of the column and cross-cut outwards through the concentric planes. These radial planes are created by thermal stresses owing to the temperature differences between the outer wall of the column and its central part.
Concentric joint sets in basic lava flows have been described in the geological literature as early as Bonney, who in 1876 referred to them as spheroidal structures and related them to the process of cooling. While this author was ahead of his time, the subsequent prevailing view presented in most fundamental geology textbooks was to relate the genesis of the ‘onion skin structure’ found in lava flows and sills to processes of weathering or deuteric alteration.
While it is not possible to set aside all previous interpretations, in many cases weathering processes or deuteric alteration act solely on planes already generated during the cooling process. Clearly, this is only one qualitative study on the shapes and genesis of these atypical structures present in basic intrusive igneous rocks; however, we feel that these structures could have important economic importance with regards to water and hydrocarbons, and therefore deserve further investigation and laboratory work.
Acknowledgments
This work is dedicated to the memory of Ken Thomson. The authors wish to thank REPSOL-YPF for permission to publish these results. Also we wish to thank J. Adamovic and an anonymous reviewer for their helpful comments and suggestions. Thanks also go to J. Soldo, P. McGregor and B. Bender, for their careful reviews, and C. Loscerbo and Mr R. Ferreira for their useful help in figure design and layout.
- © The Geological Society of London 2008
References
Concentric and radial joint systems within basic sills and their associated porosity enhancement, Neuquén Basin, Argentina
