摘要：The 100 000 m2 wave-cut pavement in theBristol Channel near Lilstock, UK, is a world-class outcrop, perfectlyexposing a very large fracture network in several thin limestone layers. Wepresent an analysis based on manual interpretation of fracture generationsin selected domains and compare it with automated fracture tracing. Ourdataset of high-resolution aerial photographs of the complete outcrop wasacquired by an unmanned aerial vehicle, using a survey altitude optimized toresolve all fractures. We map fractures and identify fracture generationsbased on abutting and overprinting criteria, and we present the fracturenetworks of five selected representative domains. Each domain is also mappedautomatically using ridge detection based on the complex shearlet transformmethod. The automatic fracture detection technique provides results close tothe manually traced fracture networks in shorter time but with a biastowards closely spaced Y over X nodes. The assignment of fractures intogenerations cannot yet be done automatically, because the fracture tracesextracted by the automatic method are segmented at the nodes, unlike themanual interpretation in which fractures are traced as a path from fracturetip to fracture tip and consist of several connected segments. This segmentation makesan interpretation of relative age impossible, because the identification ofcorrect abutting relationships requires the investigation of the completefracture trace by following a clearly defined set of rules. Generations 1 and2 are long fractures that traverse all domains. Generation 3 is onlypresent in the southwestern domains. Generation 4 follows an ENE–WSWstriking trend, is suborthogonal to generations 1 and 2, and abuts onthem and generation 3, if present. Generations 5 is the youngest fractureset with a range of orientations, creating polygonal patterns by abutting atall other fracture generations. Our mapping results show that thenortheastern domains only contain four fracture generations; thus, the fivegenerations of the outcrop identified in the southwestern domains are eithernot all present in each of the five domains or vary locally in theirgeometry, preventing the interpreter from linking the fractures to theirrespective generation over several spatially separate mapping domains.Fracture intensities differ between domains where the lowest is in the NEwith 7.3 m−1 and the highest is in the SW with 10 m−1, coinciding with different fracture orientations anddistributions of abutting relationships. Each domain has slightly differentfracture network characteristics, and greater connectivity occurs where thedevelopment of later shorter fractures is not affected by the stressshadowing of pre-existing longer fractures.
