Fracture spacing and orientation distributions for two‐dimensional data sets

Abstract

A new, automated method is developed for the measurement and display of fracture spacing and orientation from two‐dimensional (2‐D) fracture maps. Separate histograms of fracture spacing are calculated at different orientations through the 2‐D data set. The histograms for all orientations are plotted together, reducing the biasing effects associated with one‐dimensional (1‐D) sampling of a fracture population. Quantitative measurements of fracture strike anisotropy and fracture density are incorporated in the plot. No prior measurements are needed of parameters such as fracture spacing, orientation, or width. The data do not have to be preprocessed into a numerical vector format, saving time and reducing user bias. The method retains the relative frequency of different ranges of fracture spacings, which are lost in other fracture frequency measures such as fracture intensity/density. Thus the presence of any dominant (rather than average) fracture spacing and fracture clustering is highlighted. The method provides a quantitative, statistically unbiased representation of fracture orientation and spacing and is illustrated on synthetic and real fracture data sets at a range of scales (from meters to tens of kilometers). Its application to a seismic energy scattering problem is also discussed.