Abstract - Over the past decade quantitative studies of fracture systems have provided a much improved definition of the scaling properties of, and the relationships between, a broad range of fracture system attributes. These advances have arisen mainly from the increased availability of good data, enabling a more quantitative approach to data analysis that has also been stimulated by the advent of fractal concepts. Progress in the geometrical characterisation of fracture systems has been accompanied by complementary, but lesser, improvements in our understanding of the factors and processes controlling the scaling properties of fracture systems. In this talk, we provide a review of the scaling properties of a range of fracture system types, identifying some of the main geometrical characteristics of the systems and their relationship to underlying processes. We illustrate the scaling properties of fracture systems using high quality, large-scale range, datasets of fault, vein and joint systems, derived from outcrop and from 3-D seismic. We also highlight some of the shortcomings of existing data and of current analytical approaches.
A fracture system can be characterised by many geometrical attributes, some of which are independent of one another. The attributes most commonly assessed are the size distribution, density and spatial distribution of individual fracture elements, but fracture roughness and fracture connectivity have also attracted interest. Scaling systematics are most conveniently referred to in terms of two end member systems, scale-dependent and scale-independent. Fracture mode does not determine the scaling systematics, as both shear fractures and mode I extension fracture systems can be either scale-dependent or scale-independent. The primary control on scaling systematics appears to be the mechanical character of the rock volume in terms of the degree and scale of the layering. Although fault systems, for example, are often characterised by scale-independent properties, with power-law size distributions and apparently fractal spatial distributions over broad ranges of scale (e.g. more than ca 3 orders of magnitude), they can also be scale dependent. Scale dependent characteristics in fault systems occur where faulting is controlled by a given scale of layering. Even within a single fault system, scaling properties can vary from scale-independent at one scale to scale-dependent on another scale, depending on the scale of faulting relative to the scale of the mechanical layering. Other types of fracture can show similarly variable scaling systematics.
Because individual factors may operate at different scales, providing different scaling laws over different scale ranges, some caution should be exercised in interpreting scaling properties. Fault surface roughness, for example, may be scale-independent at one scale, whereas other processes such as fault segmentation and the formation of damage zones may be scale-independent over different scale ranges. Analysis of the scaling properties of fracture systems therefore requires identification of the geometrical features associated with individual processes, e.g. the formation and breaching of relay zones. Such underlying factors are best identified through an appreciation of the kinematic processes and constraints on evolution of a fracture system, rather than by statistical methods alone. Using high quality sub-surface fault datasets, for which the fault kinematics can be defined, we show how some individual fault systems have evolved from scale-independent through to scale-dependent systems, by progressive strain localisation and increased fault connectivity.
Abstract of talk given to:
Relationships Between Damage and Localized Deformation. 2nd Euroconference on Rock Physics and Mechanics, Edinburgh, November 1999