Scaling systematics of fault sizes on a large-scale range fault map

Abstract

A large-scale, high-quality fault map was produced for an intensively mined area of the U.K. South Yorkshire coalfield. This database has been used to investigate both the scaling properties of fault systems and the robustness of techniques for the prediction of sub-resolution faulting. The mapped area is 87 km 2 and contains 2257 individual fault traces with, on average, one throw reading available for every 82 m of fault trace; recorded throw values range from 10 cm to 180 m and fault trace lengths from 10 m to 12 km. Two sets of normal faults (NW- and NE-striking) are present, together with local WNW-striking dextral strike-slip zones. Size populations are power-law over at least 2.5 orders of magnitude and the exponents of the throw, maximum throw and fault trace length population curves are interrelated and, taking into account the effects of resolution on trace length populations, are internally consistent. Derivatives of the primary map reproduce resolutions typical of offshore seismics and illustrate the limitations of seismic data. Power-law fault size distributions provide a basis for sub-seismic fault prediction. For this data set, a fault throw resolution of better than 10 m would be required to provide an accurate prediction of the throw population down to ca 0.6m.