Combining topology and fractal dimension of fracture networks to characterise structural domains in thrusted limestones
Introduction
Fracturing of a rock mass is a mechanical response to an applied stress (e.g., Ramsay, 1967; Long et al., 1996), with the extent and characteristics of the resultant fracture network controlled by the mechanical properties of the rock mass, fluid characteristics, and variations in the stress field (e.g., Laubach et al., 2019). Understanding the properties and characteristics of the resultant fracture network is essential in many aspects of applied geoscience, from determining the stability of an excavation (Hoek and Brown, 1980) to identifying fluid pathways and storage volumes for minerals (Cox, 2005) or hydrocarbons (Aydin, 2000).
Fracture systems are defined as geometrical arrays of linked and often interacting fractures within a rock mass (Rouleau and Gale, 1985; Odling et al., 1999). Fracture systems have attracted much scientific attention and numerous methods have been proposed to characterise them, ranging from analysis of their kinematic behaviour, through shared and/or discrete geometry, to tectonic setting, as concisely and instructively summarised by Peacock and Sanderson (2018).
The geometric arrangements of fractures in a rock volume are typically viewed as either discrete objects in space (Barros-Galvis et al., 2015; Welch et al., 2015), or topologically, that is to say, 'in relation to one another' (Long and Witherspoon, 1985; Laubach et al., 2018), and/or in direct relation to causative mechanisms.
Studies that consider the spatial distribution of fractures as discrete objects provide valuable insights into the relationships between fractures and lithological characteristics of the fractured rock mass. For example, the spacing of fractures commonly varies with lithology or, more correctly, with differences in the mechanical properties of the lithology, such that competent lithologies display more widely-spaced fractures, for a given stress, compared to their less competent counterparts (Pollard and Fletcher, 2005; Ortega et al., 2010; Hooker et al., 2013). Fracture spacing also varies with bed thickness (Ladeira and Price, 1981) with thicker beds containing more widely-spaced fractures than their thinner equivalents, for a given stress. In folded strata, differences in the geometry of fracture patterns are related to variations in competence and bed thickness and a response to the complex strain distribution in fold systems. This results in a broad array of geometrical fracture characteristics associated with ductile/brittle-ductile fold deformation features (Cosgrove, 2015; Ferrill et al., 2016).
By contrast, topological analysis of a fracture network characterises the connectivity of the constitutive fractures in that network, rather than the inherent properties of the individual fractures (Sanderson et al., 2019). This approach has provided an improved understanding of the overall behaviour of the physical properties of the rock mass under consideration, particularly in terms of its strength, porosity, and permeability (Sanderson and Nixon, 2015).
Approaches to fracture characterisation that establish a causative relationship between a particular fracture system and the mechanism responsible for its formation require observations that can indicate a temporal link between a fracture network and the proposed process (Long et al., 1996). Examples include studies of how fracture systems of different ages (established by geochemistry) link together to control mineralisation within Archean orogenic gold (Dziggel et al., 2007) or recognition of mining-induced fractures and pre-existing geological discontinuities and how they interact to produce the rock mass around a mining stope (Grodner, 1999).
The task of relating a fracture system to a specific process is particularly challenging for rocks that have been subjected to multiple deformational events. For example, in fold-and-thrust belts deformation results from a combination of burial, changes in fluid pressure and composition, folding, thrusting, uplift and exhumation (Engelder, 1985; English and Laubach, 2017). The distribution of fractures variously reflects the different failure responses to stresses of these events due to variations in mechanical properties of the rock mass (Wennberg et al., 2006), that themselves evolve through time (Laubach et al., 2009). Progressive folding can also result in multiple generations of opening-mode fractures (Cosgrove, 2015). Consequently, polyphase deformation in fold-and-thrust belts typically results in complex, sequential overlays of fracture networks with such high abundances and intricate patterns that they are not readily described by simple fold-fault-fracture geometries (Cosgrove, 2015), or by one-dimensional descriptors (Watkins et al., 2015; Laubach et al., 2018). Fractures formed at the same time can have different orientations and mineral compositions and conversely fractures formed at different times can have the same orientations or mineralisation (Laubach et al., 2019). To properly quantify the effects of the fracture networks on the rock mass, the whole fracture system must be considered rather than apparently discrete fracture sets in a fracture network (Peacock et al., 2018).
Here we present a novel approach to the challenges involved in developing an informative, and potentially predictive, characterisation of highly fractured rock. The individual constituent fracture types within the fracture system are not separated for analysis, but rather we consider how the cumulative effects can be used to discriminate different structural domains. This approach integrates discrete topological and spatial methods for characterising fractures and fracture networks by employing fractal dimension to provide a spatial context of the distribution of the constituent fractures, and then combining those data with analyses of the observed topological relationships and interconnectivity of the fracture networks. The approach provides a more robust assessment and analysis of the fractures observed within the rock mass and their characteristics than can be achieved from application of either method in isolation. As we consider all the topological and fractal data together, all the interactions between fractures, and their effects upon the characteristics of the rock mass are defined. Moreover, this approach dramatically reduces the time taken for data collection compared to traditional fracture sampling techniques and provides large amounts of unbiased data representative of fracture network characteristics over a wide range of fracture structural domains.
We apply this technique to examine the occurrence and distribution of fracturing in well-exposed in Palaeocene limestones within the frontal thrust sheets associated with the Main Boundary Thrust (MBT) of the Himalayan fold and thrust belt (Tariq et al., 2017; Dasti et al., 2018), in a region approximately 10 km north of Islamabad, NW Pakistan (Fig. 1 and Fig. 2). Here, in a single stratigraphic unit (the Lockhart Limestone) a complex sequence of fractures can be studied across fore-thrusts, back-thrusts, and pop-up structures that all occur above, and immediately to the north of the MBT. We recognise that there are multiple generations of fractures in the study area, but as the geomechanical properties of the rock mass must be the result of all fractures combined, we contend that it is important to consider all fractures collectively to understand differences in the cumulative distribution of fracture sets related to specific structures. Restricting the structural analysis to a single stratigraphic unit removes variation in fracture characteristics related to lithology.
Section snippets
Regional geological setting
The geology of the study area, in the Potwar Basin of northern Pakistan, immediately adjacent to the capital city of Islamabad (Fig. 1), is dominated by sedimentary deposits and structural features associated with the collision of the Indian and Eurasian plates during the Himalayan Orogeny. Continual southwards-directed and décollement-related thrusting of the crust of the Indian Plate resulted in a variety of high-level fold and fault structures in the hanging walls of the major thrusts that
Nomenclature and site selection
The dataset used to test the method described in this study comprises field measurements of fractures in the limestone-dominated strata of the Lockhart Formation associated with the MBT in northern Pakistan. These strata display a spectrum of brittle geomechanical behaviours across a range of scales, within units of limestone with very low porosity, interbedded with units of argillaceous siltstone and mudstone. By restricting collection of fracture data to locations within the well-exposed
Fracture characteristics of the study sites
Four principal fracture types are observed in the limestone rocks examined in this study (Fig. 7):
- (a)
Explosive, hydro-fracture-type calcite-filled veins without any dominant orientation trends,
- (b)
unmineralised clusters of sub-parallel fractures,
- (c)
clay- or gouge-filled shear fractures typically oriented parallel to bedding or with multiple cross-cutting relationships close to folds and thrusts,
- (d)
sub-parallel, calcite-filled veins that increase in abundance with proximity to thrusts of large displacement.
Analysis of fractal and topological characteristics
In order to understand how the fracture networks vary spatially across the fold and thrust belt, the measured topological parameters and fractal dimensions are cross-correlated. The data employed to undertake this analysis are presented in Appendix 1.
Sites 1 to 6 are described in detail above as they provide an ideally oriented distribution of successive structural domains from a fore-thrust, through a pop-up to a back-thrust and the associated fracture patterns. Additional data from four
Significance of the cumulative effect of fractures
The methodology presented here is novel in that it quantifies the total rock mass of the limestone, including the fracture system within it, in a single set of measurements collected simultaneously on the fracture system. This approach not only enables efficient collection of data, dramatically reducing the time taken for data collection, but more importantly, it provides data that characterise the cumulative effects of the fractures, which may have resulted from multiple strength hardening or
Conclusions
A new approach of combining independently derived topological and fractal analyses of fracture networks has been developed to quantify the characteristics of highly deformed limestone in the Himalayan fold and thrust belt. This technique is employed to define the characteristics of complex, heterogenous fracturing in various structural settings within the hanging wall of the Himalayan Main Boundary Thrust north of Islamabad, Pakistan which has applicability to a wide variety of fracture
CRediT authorship contribution statement
Mark W. Grodner: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Visualization. Stuart M. Clarke: Conceptualization, Resources, Writing – review & editing, Supervision, Project administration, Funding acquisition. Stuart D. Burley: Conceptualization, Resources, Writing – review & editing, Supervision, Funding acquisition. A. Graham Leslie: Writing – review & editing. Richard Haslam: Writing – review & editing,
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This research is funded by Orient Petroleum Incorporated (OPI) and by the Acorn Fund at Keele University. OPI additionally provided logistical support for fieldwork in Pakistan. We gratefully acknowledge the assistance of OPI management and especially field geologists Muhammad Saleem and Israr Azfal, along with driver Khalid Nazar, for their generous assistance in the field. We thank John Walsh (Fault Analysis Group, University College Dublin) who very kindly reviewed an earlier draft of the
References (70)
Fractures, faults, and hydrocarbon entrapment, migration, and flow
Mar. Petrol. Geol.
(2000)- et al.
Fracture spacing in layered rocks: a new explanation based on the stress transition
J. Struct. Geol.
(2000) - et al.
Fault analysis and paleostress evolution in large strain regions: methodological and geological discussion of the south eastern Himalayan fold-and-thrust belt in Pakistan
J. Asian Earth Sci.
(2005) Loading paths to joint propagation during a tectonic cycle: an example from the Appalachian Plateau, USA
J. Struct. Geol.
(1985)- et al.
Advances in the implementation of the box-counting method of fractal dimension estimation
Appl. Math. Comput.
(1999) - et al.
Structural variation within the Himalayan fold and thrust belt: a case study from the Kohat-Potwar Fold Thrust Belt of Pakistan
J. Struct. Geol.
(2018) - et al.
Quantifying uncertainties in multi-scale studies of fractured reservoir analogues: implemented statistical analysis of scan line data from carbonate rocks
J. Struct. Geol.
(2010) Brittle microtectonics: principles and practice
J. Struct. Geol.
(1985)- et al.
Fracture-aperture sized frequency, spatial distribution, and growth processes in strata-bounded and non-strata-bounded fractures, Cambrian Mesón Group, NW Argentina
J. Struct. Geol.
(2013) - et al.
Fracture abundance and patterns in the Subandean fold and thrust belt, Devonian Huamampampa Formation petroleum reservoirs and outcrops, Argentina, and Bolivia
Mar. Petrol. Geol.
(2012)
Relationship between fracture spacing and bed thickness
J. Struct. Geol.
Spatial arrangement of faults and opening-mode fractures
J. Struct. Geol.
Comparison of fractal dimension calculation methods for channel bed profiles
Procedia Engineering
Circular scanlines and circular windows: new tools for characterising the geometry of fracture traces
J. Struct. Geol.
Quantifying diagenetic and stratigraphic controls on fracture intensity in platform carbonates: an example from the Sierra Madre Oriental, northeast Mexico
J. Struct. Geol.
Structural analyses and fracture network characterisation: seven pillars of wisdom
Earth Sci. Rev.
Relationships between fractures
J. Struct. Geol.
Spatial and layer-controlled variability in fracture networks
J. Struct. Geol.
Statistical characterization of the fracture system in the Stripa granite, Sweden
Int. J. Rock Mech. Min. Sci. Geomech. Abstr.
The distribution of deformation in parallel fault-related folds with migrating axial surfaces: comparison between fault-propagation and fault-bend folding
J. Struct. Geol.
The use of topology in fracture network characterisation
J. Struct. Geol.
Graph theory and the analysis of fracture networks
J. Struct. Geol.
Fractal analysis of fracture patterns using the standard box-counting technique: valid and invalid methodologies
J. Struct. Geol.
Appraisal of fracture sampling methods and a new workflow to characterise heterogeneous fracture networks at outcrop
J. Struct. Geol.
Size and spatial fractal distributions of coal fracture networks under different mining-induced stress conditions
Int. J. Rock Mech. Min. Sci.
Tectonic Evolution of the Margalla Hills and a Part of the South Eastern Hazara Ranges, Pakistan. Unpublished
Engineering classification of rock masses for the design of tunnel support
Rock Mech.
Analytical modelling and contradictions in limestone reservoirs: breccias, vugs, and fractures
Journal of Petroleum Engineering
Correcting for finite spatial scales of self-similarity when calculating the fractal dimensions of real-world structures
Proc. R. Soc. B Biol. Sci.
Fracture patterns associated with the evolution of the Teton anticline, Sawtooth Range, Montana, USA
Fractal dimension and fracture
Nature
Analysis of Reservoir Scale Fracture Sets in Southern Italy's Carbonate Reservoir Analogues, Monte Faito, Sorrento Peninsula (Southern Italy). International Petroleum Technology Conference
The association of folds and fractures and the link between folding, fracturing and fluid flow during the evolution of a fold–thrust belt: a brief review
Coupling between deformation, fluid pressures, and fluid flow in ore-producing hydrothermal systems at depth in the crust
Economic Geology 100th Anniversary
Cited by (6)
Estimation of rough fracture network permeability using fractal and topology theories
2023, Gas Science and EngineeringMapping of potential groundwater recharge sites in the Smaâla area (Central Morocco)
2023, Journal of African Earth SciencesFractal characteristics of rocks and mesoscopic fractures at different loading rates
2023, Geomechanics for Energy and the EnvironmentCitation Excerpt :Although this phase occurs randomly, it possesses statistical self-similarities. Numerous studies have demonstrated that these complex structures inside the rock and the damage characteristics after undergoing a force, are statistically self-similar at certain scales, and this property is closely related to the properties of the rock and the damage processes, which is important for studying the deformation and damage evolution laws of the rock mass under external loading.9–11 Currently, most of the conclusions regarding the mechanical properties of rocks are based on the macroscopic medium theory and do not consider the influence of the mesoscopic structure on their macroscopic phenomena.
Structural Analysis and Reservoir Characterisation of Cretaceous Sequence in Kohala Bala, Khyber Pakhtunkhwa, Pakistan
2022, Rudarsko Geolosko Naftni ZbornikEffect of Heterogeneity on the Extension of Ubiquitiformal Cracks in Rock Materials
2022, Fractal and Fractional