Shallow-level igneous intrusions are a common feature of
many sedimentary basins, and there is increased recognition of the
syn-emplacement deformation structures in the host rock that help to
accommodate this magma addition. However, the sub-seismic structure and
reservoir-scale implications of igneous intrusions remain poorly understood.
The Trachyte Mesa intrusion is a small (
Syn-emplacement deformation structures (faults, fractures and forced folds) provide the principal mechanism for the accommodation of magma at shallow crustal levels (e.g. Pollard and Johnson, 1973; Hansen and Cartwright, 2006; Senger et al., 2015; Wilson et al., 2016). The thermal effects of intrusions on host rocks have been well studied (e.g. Jaeger, 1964; Irvine, 1970; Brooks Hanson, 1995; Rodriguez-Monreal et al., 2009; Senger et al., 2014; Aarnes et al., 2015; Gardiner et al., 2019), as too has been the hydrothermal fluid flow associated with their emplacement (e.g. Rossetti et al., 2007; Scott et al., 2015). However, the impact of these syn-emplacement deformation structures on post-emplacement fluid flow around intrusions is less well understood (Montanari et al., 2017). Our study is based on detailed kinematic, geometric and spatial analyses of networks of deformation bands in a host-rock sandstone associated with emplacement of the Trachyte Mesa igneous intrusion, Henry Mountains, Utah (Fig. 1). The aims are to: (1) characterize deformation structures developed in the Entrada Sandstone around the intrusion; (2) track geometrical and topological changes towards the intrusion; and (3) discuss the implications of these changes for fluid flow in the host-rock sandstone.
Geological setting and study area.
Deformation bands are a common structure in many fluid and gas reservoir sandstones. In particular, weakly cemented and highly porous sandstones are ideal candidates for the development of deformation bands (e.g. Aydin, 1978; Underhill and Woodcock, 1987; Shipton and Cowie, 2001; Fossen et al., 2007; Wibberley et al., 2007). Numerous studies have shown that deformation bands can have a significant influence on fluid flow (Antonellini and Aydin, 1994, 1995; Gibson, 1998; Sigda et al., 1999; Fossen and Bale, 2007; Ballas et al., 2015). Due to their mode of formation (i.e. mainly cataclasis and compaction; Du Bernard et al., 2002; Aydin et al., 2006; Fossen et al., 2007; Eichhubl et al., 2010) deformation bands tend to have lower permeabilities than their host sandstones and, in turn, they negatively affect fluid flow (Sternlof et al., 2004; Fossen and Bale, 2007; Rotevatn et al., 2013). Porosity and permeability reductions due to deformation bands may significantly reduce reservoir connectivity by creating baffles to fluid flow (e.g. Taylor and Pollard, 2000; Sternlof et al., 2006; Torabi et al., 2008; Sun et al., 2011; Saillet and Wibberly, 2013; Skurtveit et al., 2015) and even, in some cases, act as seals to hydrocarbon accumulations (e.g. Knipe et al., 1997; Ogilvie and Glover, 2001; Parnell et al., 2004).
Deformation bands can develop in most structural and tectonic settings, provided the host rock is susceptible to their formation (Fossen et al., 2007; Schultz et al., 2010; Soliva et al., 2013; Ballas et al., 2015). Deformation bands within quartz arenite to arkosic sandstones (i.e. those lacking lithics) preferentially form in weekly cemented layers at shallow depths (1–3 km; Fossen, 2010). This depth regime is coincident with the emplacement of many shallow-level igneous intrusions, and deformation bands have been reported to develop as accommodation structures associated with sills and laccoliths (e.g. Morgan et al., 2008; Wilson et al., 2016, 2019; Westerman et al., 2017; Fig. 1). These deformation bands may have important implications for compartmentalisation and fluid flow within reservoirs hosting intrusions. To date, few quantitative analyses have been carried out to analyse the deformation structures associated with movement and accommodation of mobile substrates such as salt (e.g. Antonellini and Aydin, 1995) or magmatic intrusions (e.g. Morgan et al., 2008; Senger et al., 2015).
A fracture network can be regarded as a system of fractures (including deformation bands) developed within the same rock volume and may be made up of multiple fracture sets (e.g. Fig. 2; Adler and Thovert, 1999; Peacock et al., 2016). Fractures are generally described by their geometry (e.g. orientation and length) and characteristic attributes (e.g. fracture type, morphology and mineral fill). These attributes may be used to define fracture sets (Priest, 1993; Adler and Thovert, 1999; Sanderson and Nixon, 2015; Procter and Sanderson, 2018), which are often used to delineate distinct structural events within the evolution of a wider fracture network. Wilson et al. (2016) used the term “phases” to describe what are in effect the various fracture sets observed in the Trachyte Mesa study area. Through these attributes it was possible to gain a good understanding of the various fracture sets and networks, however the relationships between these systems (e.g. connectivity) requires further analysis. Sanderson and Nixon (2015, 2018) highlighted the use of “topology” for describing the relationships between geometrical objects and, building on the work of Mauldon (2001), Manzocchi (2002), Rohrbaugh et al. (2002) and Mäkel (2007), outlined a workflow for fracture analysis. Such topological analysis of fractures can be extended to other discrete structures and in this case has been applied to deformation bands as part of the present study.
Sampling traverse across lateral margin of the Trachyte Mesa
intrusion.
The Trachyte Mesa intrusion is a small (1.5 km
Trachyte Mesa has an elongate, laccolithic geometry, trending NE–SW (Fig. 1b) with a thickness varying from 5–50 m (Morgan et al., 2008). The
intrusion formed by the amalgamation and stacking of multiple thin
(
The Entrada Sandstone Formation (part of the San Rafael Group) is Jurassic (Callovian) in age and is composed of a mixture of white cross-bedded sandstones, reddish-brown silty sandstones, siltstones, and shale beds (Aydin, 1978; Fig. 2a). The Entrada was deposited in an aeolian environment and extends over a vast area, making it the largest of the Colorado Plateau ergs (Hintze and Kowallis, 2009). Entrada Sandstone is generally quartz-dominated (Aydin, 1978), although a subarkosic composition for rock units studied around the Trachyte Mesa intrusion may be a more appropriate lithological description. Calcite is the most common cement, although siliceous and pelitic cements are abundant in some layers (Aydin, 1978).
The Entrada Sandstone, being highly porous, is the ideal lithology for the
formation of deformation bands (Fig. 2) and, as a result (along with the
Lower Jurassic Navajo and Wingate sandstones, also found on the Colorado
Plateau and stratigraphically below the Entrada; Jackson and Pollard, 1988),
has been the focus of several studies on such structures (Aydin, 1978; Aydin
and Johnson, 1978, 1983; Shipton and Cowie, 2001; Fossen and Bale, 2007;
Fossen et al., 2007). These Jurassic sandstones form natural fluid carrier
systems and reservoirs for hydrocarbon and CO
Deformation bands in the vicinity of Trachyte Mesa generally form as
conjugate sets striking roughly NE–SW and individual bands are generally
narrow (
Fracture analysis of hand specimens.
Outcrop studies and rock samples were collected from a structural transect
across the north-western lateral margin of the Trachyte Mesa intrusion
(Figs. 2 and 3), in order to quantify the change in the network observed
across the intrusion margin. The study consists of six structural stations,
relatively evenly spaced, from
As roughly NE–SW trending deformation bands are the dominant structural orientation identified along the margin (Fig.1; Wilson et al., 2016), care was taken to ensure that the surfaces photographed, and subsequently analysed, were oriented in a similar, optimal perpendicular (NW–SE) orientation in order to sample the network most appropriately. Due to this sampling technique, results will only be appropriate for analysing fluid flow across (perpendicular to) the intrusion margin, and further analysis may be necessary to understand flow parallel to the intrusion margin. It is acknowledged that by only carrying out studies in one orientation we are invoking an orientation bias into our results. However, choosing sections at a high angle to the main orientation of the band intersections should minimise the bias in both the geometrical and topological parameters.
A selection of rock samples was collected at each structural station (Fig. 3) in order to carry out hand-specimen and thin-section (i.e. petrological, porosity and microstructural) studies. Samples were oriented in the field in order to enable thin sectioning in a similar vertical, NW–SE oriented plane to the outcrop photograph/scan surfaces. Ensuring that similarly orientated sample areas are studied at all scales increases the chances of sampling the same fracture systems (i.e. NE–SW trending fracture networks) as observed in outcrop, and thus the resulting scalar statistics should be more appropriate.
Various analytical techniques have been proposed for the investigation of fracture networks (e.g. Walsh and Watterson, 1993; Berkowitz, 1995; Adler and Thovert, 1999). In this study the methods outlined by Sanderson and Nixon (2015) and Procter and Sanderson (2018) have been applied. The method was described in detail by Sanderson and Nixon (2015), and only a brief summary is given here. The basic principle is outlined in Fig. 4, and comprises the mapping of a 2D fracture network, measuring fracture (or branch) lengths and quantities, and node counting (e.g. Fig. 5).
Schematic image outlining the principal method applied for fracture analysis (Sanderson and Nixon, 2015). Branches and nodes are shown on fracture trace (A–B): I-nodes (red circles); Y-nodes (blue triangles); X-nodes (green squares). Proportions of I-, Y- and X-nodes may be plotted on a ternary plot to visualise different fracture network types (after Manzocchi, 2002).
Example of fracture analyses undertaken at different scales in
this study, sample TMFS-5.
Using the outcrop photographs, deformation band networks were mapped to the highest level of detail attainable from the image resolution. Areas were then selected in order to sample the networks at each site. In areas of more heterogeneous deformation, multiple areas were sampled at different scales (ranging from 20–100 cm diameter circles) in order to capture the variability. Circular scan-lines/areas were used rather than squares as these provide the least orientation bias, with an equal likelihood of sampling any given fracture orientation on a 2D surface (Mauldon, 1994; Rohrbaugh et al., 2002; Procter and Sanderson, 2018).
For each sample circle, the total number of deformation bands and total band length were recorded. In addition, the total number of branches (i.e. segments between intersecting deformation band points or nodes) was also determined. Sanderson and Nixon (2015) argued that preference should be given to the use of branches as it is often difficult to recognise an individual, continuous fracture trace within a deformation band network, whereas branches are uniquely identifiable. Furthermore, as exposures and sampling areas are of finite size, many deformation bands extend beyond the sample area. Therefore, the frequency and length of deformation bands will be subject to sampling bias (Riley, 2005). By contrast, the length of branch lines is likely to be less censored, thus reducing this sampling bias issue.
Using sample area, total number of deformation bands (or branches) and total deformation band (branch) length, a number of fracture network characteristics can be defined. These include: frequency (total number/area); intensity (total length/area); spacing (the inverse of intensity, i.e. area/total length); characteristic fracture length (mean length; total fracture length/ total number of deformation bands); and dimensionless intensity (multiplying fracture intensity by the characteristic length). Details on the derivation of these terms were provided by Sanderson and Nixon (2015).
A given deformation band network consists of lines, nodes and branches
(Figs. 4 and 5a–b). As outlined above, lines will consist of one or more
branches, with nodes (i.e. fracture intersections) at either end of each
branch. Three main types of nodes exist: I-nodes (isolated fracture
terminations within the host rock); Y-nodes (where one fracture terminates
against another); and X-nodes (where two deformation bands cross-cut one
another). Within a sample area, a fourth type of node may also be recorded,
where deformation bands intersect the outer perimeter of the sample area
(termed E-nodes; Sanderson and Nixon, 2015). As discussed by Procter and
Sanderson (2018), combining node counting with a measurement of intensity
(usually
The proportion of I-, Y- and X-nodes have been used by various authors to characterize a fracture network (e.g. Manzocchi, 2002; Mäkel, 2007) and the results plotted on a triangular diagram (Fig. 4). As the relative proportion of nodes will remain unchanged by any continuous transformations (i.e. scale changes and strains), this is termed a topological classification (Sanderson and Nixon, 2015).
Optical microscopy petrographical, porosity and microstructural analyses were carried out on thin sections cut from each hand specimen (e.g. Fig. 5c). Sections were impregnated with blue-dyed plastic resin in order to highlight porosity. Both compositional percentages were visually estimated using percentage estimation comparison charts (Bacelle and Bosellini, 1965; Tucker, 2001), while quantitative analyses of porosity percentages were attained using the image analysis software package ImageJ (Fig. 6; Schneider et al., 2012; Heilbronner and Barrett, 2014).
Quantitative determination of porosity percentages using ImageJ.
Six structural stations were analysed across an approximately 100 m-long transect over the north-western intrusion margin (Fig. 2 and Appendix). In order to maximise the area sampled at each scan station, data from multiple scan circles have been aggregated. Note, where scan circles overlap, data have been omitted from the totals to avoid duplication, e.g. at Station 1 where all smaller scan circles lie within the larger circle. Results are summarised in Table 1 and Fig. 7. The values for Station 2, and to a lesser extent Station 1, are based on relatively few measurements. Procter and Sanderson (2018) recommended at least 30 nodes, so the data from Station 2 (7 nodes) should be considered unreliable, compared to the other stations where the number of nodes varies from 24 (Station 1) to 847 (Station 5).
Summary table showing total values for each structural station. Total values do not include hand specimens due to their small size making estimates of frequency and intensity unreliable, though trends for samples do match those for outcrops in this study. For details on individual scan circles see the Appendix.
Nodal and fracture analysis results.
Nodal populations for each station were recorded (Table 1, Fig. 7b) and plotted on triangular diagrams (Fig. 7c; after Manzocchi, 2002). The outcrops studied show a clear dominance of I- and Y- nodes (Fig. 7), although the proportion of these nodes varies across the transect (Fig. 7b). Structural stations more distal to the intrusion (i.e. Stations 1 and 2) show approximately equal proportions of I- and Y-nodes (Fig. 7). The proportion of I-nodes decreases through Stations 3 to 4, where Y-nodes become dominant with proximity to the intrusion. At Stations 5 and 6 overlying the intrusion, I-nodes are negligible, and the nodal populations are dominated by Y- and X-nodes. These results reflect the overall increase in connectivity of the conjugate deformation bands observed at the intrusion margin (Morgan et al., 2008; Wilson et al., 2016).
The abundance of deformation bands increases with proximity to the intrusion
(Fig. 7d). The frequency of branches (
The change in node types indicates a change in topology across the transect
(Fig. 7b). Stations 1 and 2 plot in the upper region of the I-Y-X triangle
(Fig. 7c), which is where “tree-like” networks, with few branches
enclosing blocks typically develop (e.g. Fig. 2b, c), whereas the other
stations are more dominated by connected nodes (Y and X), typical of
networks with lots of branches and deformation bands that enclose many small
blocks (as seen in Figs. 3e, f and 5a, b). These topological changes can be
monitored by several key parameters (Fig. 7f). The connections per line
(
Thin section analysis of outcrop samples further shows a significant
increase in deformation and fracture intensity with proximity to the lateral
intrusion margin. Outboard (
Porosity variability across the intrusion margin.
Thin section photomicrographs showing host rock composition
structure and porosity. Note overall decrease in porosity (blue dye
staining) and increase in deformation from sample TMFS-1 through to TMFS-5.
TMFS-2 appears to sample a slightly coarser-grained bed within the tracked
sandstone unit (Fig. 2). Laminations are clearly visible at both the
hand-specimen and thin-section scale (Figs. 3b and 8a). Thin-section
analysis shows a well-sorted sample with similar subarkose composition to
TMFS-1. Sub-rounded grains suggest that this sandstone is relatively mature.
Large patches of poikilotopic sparry calcite are again present (Fig. 9c).
Similar to TMFS-1, this sample also shows no visible deformation bands.
TMFS-2 exhibits similar porosities to that of TMFS-1, with an average
porosity of
Approaching the lateral intrusion margin (
Thin section photomicrographs showing deformation band structure
and porosity (edges of deformation bands highlighted in by red markers at
edges of image).
Moving onto the intrusion margin, sample TMFS-4, background (host rock)
porosity is variable from lamina to lamina, although the average remains
relatively high at 14.4 % (Figs. 8, 9e). The host rock shows evidence for
strain, with grains exhibiting embayed contacts, intragranular microcracking
and transgranular fractures (Fig. 9e). TMFS-4 samples three discrete
(
Immediately above the lateral intrusion margin, sample TMFS-5 (Fig. 8a)
displays significant deformation zones. Background (host rock) porosity is
lower than the less deformed samples described above (average 7.0 %; Figs. 8b, 9f). This is due to greater compaction, as evidenced by the higher
proportion of interlocking (more tightly packed) grains, embayed contacts
and possible pressure solution with sutured grain contacts and intragranular
fractures (Fig. 9). Calcite cementation within the background rock is
patchy, with calcite spar accounting for only 2 %–3 % of porosity
reduction. Sampling several deformation bands, these appear more diffuse (up
to 1 cm wide; Figs. 8a and 10c–d) than those sampled in TMFS-4. Although not
well-established, distinct slip zones may be identified within deformation
bands. Microporosity within the deformation bands is extremely low
(
Passing over the upper hinge zone of the monoclinal intrusion margin, sample
TMFS-6 (Fig. 8a) shows a clear system of moderately-dipping
cross-laminations. Again, a subarkose composition is apparent (quartz grains
with lesser plagioclase- and microcline-feldspar). Embayed contacts are
visible showing pressure solution/dissolution and focused intragranular
fractures (e.g. Fig. 10e). Background porosity is significantly reduced in
TMFS-6 compared to the other 5 samples, at 3 % to 9 % (average porosity 6.1 %; Fig. 8b). This is largely due to a combination of compaction,
cataclasis and greater calcite cementation (Fig. 10e–h). Multiple diffuse
and discrete, anastomosing deformation bands are identifiable (Figs. 8a and
10e–h). Microporosity within the deformation bands is
A clear increase in fracture abundance can be observed across the intrusion margin. The quantitative analysis of fracture attributes, such as intensity, frequency and dimensionless intensity, increase progressively across the intrusion margin (Fig. 7). In addition, analysis of nodal populations highlights the topological change across this same area. As fracture frequency and intensity increase onto the intrusion, this is accompanied by an increase in Y (and to a lesser extent X) nodes. Manzocchi (2002) and Sanderson and Nixon (2018) both linked changes in nodal populations to critical dimensionless intensity, and percolation thresholds, using stochastic models.
Figure 7 shows the nodal distributions for this study overlain on contoured
triplots defining lines of critical branch dimensionless intensity
(
Nodal populations were also acquired at the hand-specimen scale (Fig. 3 and
Appendix). Similar topological trends are apparent from these hand-specimen
samples; however, values appear more extreme at both end members (i.e.
Figure 11 shows schematic 3D block diagrams which compare the distribution of deformation band structures across the Trachyte Mesa intrusion to forced folds above a normal fault (Ameen, 1990; Cosgrove, 2015). The variations in the deformation band network geometry seen across the Trachyte Mesa intrusion margin (Wilson et al., 2016) are very similar to those in the model for forced folds above a normal fault. The increase in fracture intensity, frequency and dimensionless intensity is also consistent with this model, with deformation increasing across the forced fold. Offsets are dominantly extensional, consistent with the forced-fold model.
Schematic 3D block diagrams and cross sections comparing the
distribution of deformation structures.
The analogy to the growth of a normal fault is viable due to the mode of emplacement of the Trachyte Mesa intrusion through vertical stacking of sill sheets (Morgan, 2008; Wilson et al., 2016), which represent the uplifted footwall block. As the intrusion grows in size (by the incremental addition of sill sheets) this drives the shear localisation of deformation similar to that of a propagating normal fault (e.g. Ballas et al., 2015). The model assumes a two-phase growth mechanism for individual sheets, whereby the sill sheet propagates laterally as a thin layer and then vertically inflates (Hunt, 1953; Corry, 1988; Wilson et al., 2016). The vertical inflation phase of each individual sheet would therefore mimic individual fault growth/slip events on a normal fault scarp. However, as highlighted by Wilson et al. (2016), the order of stacking of sill sheets (over-, under-, mid-accretion; Menand, 2008) can significantly impact the style of syn-emplacement deformation within the overlying host-rocks (Fig. 11c). The transect in this study samples a section of the intrusion margin which displays out-of-sequence (i.e. under- and mid-accretion; Menand, 2008) stacking, which leads to a broader monoclinal margin. In contrast, in a stepped margin (resulting from over-accretion of sill sheets), a more complex zonal variability in the fracture network and topology may be observed, rather than the gradual change seen for the monoclinal margin in this study.
Due to a lack of Entrada host rock exposures across the top surface of the intrusion, the transect in this study does not extend a significant distance onto the intrusion top surface, in order to sample the deformation style and intensity away from the monoclinal margin. There are, however, a limited number of host rock exposures distributed across the wider intrusion top surface. These do not appear to exhibit the intensity of deformation bands observed in this study, though sandstone porosities do appear reduced (based on field observations at outcrop, but not sampled), suggesting that the style of deformation over the intrusion top surface differs markedly from that along the intrusion margin (e.g. Fig. 9b).
Microstructural analysis of deformed samples shows dominantly brittle
deformation with cataclastic flow and compaction occurring within
deformation bands. Despite significant porosity reduction from undeformed
host rock (typically 10 %–23 %) to within deformation bands,
micro-porosity of
Deformation bands outboard of the intrusion margin (i.e. TMFS-1 to -3) appear to show dominantly compaction related deformation, with minor cataclasis (Figs. 8 and 9). These would therefore be best categorised as pure and/or shear-enhanced compaction bands (PCBs and SECBs; Eichhubl et al., 2010; Ballas et al., 2015). As you move closer to the intrusion margin, more embayed contacts and evidence for pressure solution are observed (e.g. TMFS-4; Fig. 9). This may be an indication of shortening and an increasing margin perpendicular stress with proximity to the intrusion. Deformation bands in samples collected from localities above the intrusion (i.e. TMFS-5 and -6) show significantly more evidence for cataclasis and grain shearing (Fig. 10), highlighting the strain localisation in this domain (Fig. 11b). Although strain localisation is evident, none of the deformation bands analysed in this study exhibit well-defined principal slip surfaces or fault cores; such deformation band fault zones are, however, more common in areas of the intrusion margin where sill sheets are stacked in a normal sequence and where strain is localised at individual sill-tip terminations (Fig. 11c; Wilson et al., 2016).
As discussed by Ballas et al. (2015), these different deformation band types may each have subtle differences in how they impact the overall permeability and flow pathways within the sandstone. PCBs and SECBs may reduce the local permeability by two orders of magnitude, however the lack of cataclasis may negate these bands from forming barriers to flow (Rotevatn et al., 2013), but may influence flow pathways (Sternlof et al., 2006). In contrast, cataclastic bands will also reduce the local permeability by two, or more, orders of magnitude (Ballas et al., 2015), but may also significantly impede flow due to additional fabric anisotropy, forming barriers (e.g. Ogilvie et al., 2001) and significantly impacting flow pathways (Taylor and Pollard, 2000; Soliva et al., 2013). Therefore, in addition to the topological variations outlined in Fig. 7, understanding the deformation band type is also an important consideration.
In addition to porosity reduction due to deformation bands, a reduction in host-rock porosity is apparent within samples TMFS-5 and TMFS-6, from sandstone beds overlying the intrusion (9 %–23 % in samples TMFS-1 to -3 compared to 4 %–9 % in samples TMFS-5 and -6). This reduction appears to be the result of greater compaction of grains and an increase in cementation. The increased cementation observed in samples TMFS-5 and -6 (as well as the presence of calcite spar in the more distal outcrops) could be related to the circulation of warm fluids around the intrusion during emplacement. As the solubility of calcite decreases with increasing temperature, the heat introduced by the intrusion could facilitate precipitation of calcite from surrounding groundwater of appropriate composition. This is consistent with the observed porosity reduction and thinning of beds over the monoclinal intrusion margin observed by Morgan et al. (2008).
In addition to these various host rock deformation structures impacting fluid flow, of course by far the most significant impact on the reservoir scale permeability framework is the intrusion itself. Permeability within the intrusion is extremely low and so regional fluid flow pathways will first be influenced by the easiest route around the intrusion, which will be influenced not only by the distribution and connectivity of deformation structures discussed in this study, but also by the permeability of the surrounding undeformed host rock.
Shallow-level intrusions are a common feature of many hydrocarbon basins, including: the NE Atlantic margin (e.g. Malthe-Sørrenssen et al., 2004; Hansen and Cartwright 2006); West of Shetland (e.g. Rateau et al., 2013; Gardiner et al., 2019); and the southern and north-western margins of Australia (e.g. Holford et al., 2012; Mark et al., 2020). The present quantitative study of deformation bands highlights the significant impact magma emplacement can have in highly porous siliciclastic reservoir systems. Although only a small study, results show that deformation band abundance and intensity increase markedly across the NW margin of the Trachyte Mesa intrusion. The methods applied provide a means of quantifying this increase in deformation intensity across an intrusive margin.
The deformation bands show significant porosity reduction that is most apparent in the sandstones overlying the intrusion. The overall porosity reduction demonstrated in Fig. 8 would produce approximately an order of magnitude change in permeability (e.g. assuming Kozeny-Carman equation fundamentals; Civan, 2002, 2015), as observed in many reservoir rocks. However, this assumes a homogeneous development of the grain-scale processes (i.e. grain size and sphericity), and so the heterogeneity of deformation bands make the application of the Kozeny-Carman equation an oversimplification. Microstructural analysis suggests that the porosity reduction is largely through localized development of deformation bands. These have been shown to start away from the intrusion as poorly connected (or unconnected) networks, which might baffle and reduce fluid flow, but probably to no great significance. In comparison, in the host rocks above the intrusion margin, the increase in Y- and X- nodes highlights the significant increase in deformation band interconnectivity, which in turn will significantly reduce the network connectivity and permeability pathways of the sandstone. Importantly, the formation of a connected network of such bands may reduce permeability by several orders of magnitude (e.g. Ballas et al., 2015).
The deformation aureole immediately bordering the intrusions has not been
analysed as part of this study. However, this is an important factor to
consider when assessing the likely impact that intrusion-related deformation
may have on a wider reservoir system. At Trachyte Mesa, deformation bands
appear to decrease markedly from
Although additional analyses are required in order to understand the 3D
connectivity of these fracture systems, the present 2D analytical study goes
a long way to establishing the connectivity of deformation bands in the host
rocks to the Trachyte Mesa intrusion. The more pertinent issue is
understanding the effects this connectivity could have on permeability
within the host rocks. This study emphasises the potential importance of
understanding the impact of syn-emplacement deformation to localised fluid
flow around igneous intrusions. Gaining a better understanding of these
emplacement-related deformation structures may have important implications
for fluid flow, hydrocarbon reservoir connectivity/deliverability,
hydrology, geothermal energy and CO
Deformation structures vary in style and intensity across the lateral “monoclinal” margin of the Trachyte Mesa intrusion, but there is a clear relationship between deformation and proximity to the intrusion margin. This has led a number of authors to propose that these deformation structures developed in response to emplacement of the intrusion (e.g. Johnson and Pollard, 1973; Morgan et al., 2008; Wilson et al., 2016; this study).
Although only a small study, our results show that deformation bands increase in abundance and intensity across the NW margin of the Trachyte Mesa intrusion. The methods applied provide a means of quantifying this increase in deformation intensity across the intrusive margin. Furthermore, the application of topologic analysis (in the form of nodal analysis) provides a means of understanding the network connectivity of deformation structures, and thus their negative impact on reservoir permeability. The increase in margin parallel Y- and X-nodes with proximity to the intrusion is likely to inhibit flow perpendicular to the intrusion margin, as well as potentially forming non-producible reservoir units.
This study highlights that fluid flow in deformed host rocks around igneous bodies may vary significantly from that of the undeformed host-rock reservoir. Therefore, a better understanding of the variability of deformation structures, and their association with intrusion geometry, will have important implications for industries where fluid flow within naturally fractured reservoirs adds value (e.g. hydrocarbon reservoir deliverability, hydrology, geothermal energy and carbon sequestration).
Sample station TMFS-1 scan circles
Sample station TMFS-2 scan circles
Sample station TMFS-3 scan circles
Sample station TMFS-4 scan circles
Sample station TMFS-5 scan circles
Sample station TMFS-6 scan circles
The data set used in this paper can be found in the Supplement.
The supplement related to this article is available online at:
PIRW designed and conducted the research, interpreted the data and prepared the manuscript. RWW designed the study and assisted with data analysis and manuscript preparation. DJS assisted with data analysis and manuscript preparation. IJ assisted with data analysis and manuscript preparation. KJWM assisted with manuscript preparation.
The authors declare that they have no conflict of interest.
This article is part of the special issue “Faults, fractures, and fluid flow in the shallow crust”. It is not associated with a conference.
The authors would like to thank from Laurel Goodwin, Craig Magee, Peter Eichhubl and an anonymous reviewer for their constructive reviews and editorial feedback which have helped enhance the manuscript. The authors would also like to thank Casey Nixon for advice during the development of the work. Penelope I. R. Wilson acknowledges Kingston University London for PhD funding and laboratory access that supported this research. David J. Sanderson acknowledges support from a Leverhulme Emeritus fellowship during the development of this work.
This paper was edited by Peter Eichhubl and reviewed by Laurel Goodwin and one anonymous referee.